State determination method, state determination device, state determination system, and learning method of farm field

Abstract

To provide a state determination method, a state determination device, a state determination system, and a learning method by which the state of a farm field can be estimated.SOLUTION: A state determination method of a farm field executed by an arithmetic circuit includes: a step of acquiring the content of gas measured in surface layer soil of soil in a farm field or the soil; and a step of estimating the degree of microorganism activity in the soil from the acquired gas content on the basis of the relationship between a previously prepared gas content and the degree of microorganism activity.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to a condition determination method, a condition determination device, a condition determination system, and a learning method for determining the condition of a farm field by measuring gas in the farm field. [Background technology]

[0002] With the aim of improving the yield or disease resistance of plants and agricultural products, technologies that analyze the growth conditions of plants and agricultural products in fields and environmental information to determine the amounts and timing of spraying water, chemical fertilizers, pesticides, etc. are becoming more widespread. For example, Patent Document 1 discloses acquiring field data, which is information related to reflected light obtained by irradiating a field with near-infrared light from a flying drone, and acquiring meteorological information including the amount of solar radiation, temperature, moisture content of the atmospheric layer, humidity, and carbon dioxide concentration.

[0003] In recent years, the extraction and cultivation of specific microorganisms that can support the natural growth of plants and agricultural crops has been technically established. As a result, from the perspective of the environment and biodiversity, agricultural methods that spray specific microorganisms are being realized in various forms as agricultural methods that can improve yields or disease resistance without using pesticides or chemical fertilizers (e.g., Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7075127 [Patent Document 2] Japanese Patent Publication No. 2022-083572 Summary of the Invention [Problem to be solved by the invention]

[0005] In farming methods that spray specific microorganisms, the contribution of microbial activity to the growth of plants and crops is greater than in conventional farming methods, and the amount of microorganisms to be sprayed must be determined according to the activity of the microorganisms. However, with techniques such as those described in Patent Document 1, it was difficult to determine the condition of the field in order to determine the amount of microorganisms to be sprayed according to the activity of the microorganisms.

[0006] An object of the present disclosure is to provide a state determination method, a state determination device, a state determination system, and a learning method that can estimate the state of a farm field. [Means for solving the problem]

[0007] A method for determining the state of a farm field executed by an arithmetic circuit according to one aspect of the present disclosure includes the steps of acquiring the amount of gas present in the surface soil of the farm field or measured in the soil, and estimating the microbial activity in the soil from the acquired amount of gas present based on a relationship between the amount of gas present and the microbial activity prepared in advance.

[0008] A condition determination system according to one aspect of the present disclosure includes a placement device including a sensor that measures the amount of gas present in the surface soil or in the soil in a field, a placement mechanism that can place the sensor in the soil, and a mobile device that can move around the field, a condition determination device, a cleaning device that can remove dirt adhering to the sensor, and a sensor management device that includes a calibration device that can calibrate the sensor.

[0009] A method for learning an estimation model executed by an arithmetic circuit according to one aspect of the present disclosure includes the steps of acquiring the amount of gas present in the surface soil of the soil and the gas measured in the soil, acquiring the microbial activity of the soil, and learning an estimation model that estimates the microbial activity from the amount of gas present by learning using training data that includes the acquired microbial activity as input data and the acquired amount of gas present as output data. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to provide a state determination method, a state determination device, a state determination system, and a learning method that can estimate the state of a farm field. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a schematic diagram of a state determination system according to a first embodiment. [Figure 2] 10 is a flowchart showing an example of an estimation model generation process executed by an arithmetic circuit. [Figure 3] 1 is a graph showing an example of the relationship between gas concentration and microbial activity in soil. [Figure 4] 10 is a flowchart showing an example of a state determination process executed by an arithmetic circuit. [Figure 5] Graph showing an example of the relationship between microbial activity and increased abundance in soil. [Figure 6] FIG. 10 is a diagram showing an example of a heat map in which the microbial activity estimated in a farm field is mapped. [Figure 7] FIG. 10 is a diagram showing an example of a heat map that maps the amount of microorganisms applied to a farm field. [Figure 8] FIG. 10 is a schematic diagram of a state determination system according to a second embodiment. [Figure 9] 10 is a flowchart showing an example of an estimation model generation process executed by an arithmetic circuit in the second embodiment. [Figure 10] 1 is a graph showing an example of the relationship between gas concentration, temperature, and microbial activity in soil. [Figure 11] 10 is a flowchart showing an example of a state determination process executed by an arithmetic circuit in the second embodiment. [Figure 12] Graph showing an example of the relationship between microbial activity and increased abundance in soil. [Figure 13] FIG. 10 is a schematic diagram of a state determination system according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. However, the configurations described below are merely examples of the present disclosure, and the present disclosure is not limited to the following embodiments. The technology in the present disclosure is not limited to these embodiments, and various modifications, substitutions, additions, omissions, etc. are possible depending on the design, etc., as long as they do not deviate from the technical concept of the present disclosure.

[0013] Although the present disclosure has been fully described in connection with the preferred embodiments with reference to the accompanying drawings, various changes and modifications will be apparent to those skilled in the art, and such changes and modifications are to be understood as included within the scope of the present disclosure as defined by the appended claims unless they depart therefrom.

[0014] In this disclosure, when multiple embodiments are described, differences from embodiment 1 will be mainly described. In this case, components in the other embodiments that are the same as or equivalent to those in embodiment 1 will be described using the same reference numerals. Also, in the other embodiments, descriptions that overlap with embodiment 1 may be omitted.

[0015] Conventionally, the extraction and cultivation of specific microorganisms that can support the natural growth of plants and agricultural crops have been technically established. Furthermore, agricultural methods that improve the yield or disease resistance of agricultural crops by spraying microorganisms on fields are being realized. However, it has been difficult to determine the amount of microorganisms to be sprayed based on the condition of each soil in the field.

[0016] A field condition determination method executed by a calculation circuit according to the present disclosure includes a step of acquiring the amount of gas present in the surface layer of soil or in the soil in the field. The condition determination method further includes a step of estimating the microbial activity in the soil from the acquired amount of gas present based on a predetermined relationship between the amount of gas present and the microbial activity. In this way, the calculation circuit can determine the condition of the soil by estimating the microbial activity in the soil from the amount of gas present in the soil in the field. According to the condition determination method according to the present disclosure, measuring gas in the field can improve the accuracy of estimating the state of microorganisms in the soil, which greatly contributes to promoting the growth of plants and crops. Furthermore, the calculation circuit can determine the amount of microorganisms to be sprayed appropriate for the soil based on the soil condition. Therefore, according to the condition determination method according to the present disclosure, an appropriate amount of microorganisms can be determined depending on the soil condition, allowing a user of the method to efficiently promote the growth of plants and crops by spraying the appropriate amount of microorganisms on the soil.

[0017] (Embodiment 1) [1-1.Configuration] A state determination system 1 according to a first embodiment of the present disclosure will be described with reference to Fig. 1. Fig. 1 is a schematic diagram of the state determination system 1 according to the first embodiment of the present disclosure.

[0018] The condition determination system 1 comprises a sensing unit 2 including a detection device 10 and a control device 20, a placement device 30, and a management device 40. The detection device 10 has one or more sensors and functions as a detection unit that detects various information in the condition determination system 1. The control device 20 functions as a control unit that determines the condition of the field based on information acquired from the detection device 10. The placement device 30 functions as a transport unit that moves and places the detection device 10 in a predetermined position. The management device 40 functions as a sensor management unit that manages the one or more sensors provided in the detection device 10 and stably maintains the quality of the sensors.

[0019] In the first embodiment, the detection device 10 includes a gas sensor 11. The detection device 10 may include one gas sensor 11 for measuring one gas, or multiple gas sensors 11 for measuring multiple gases. The gas sensor 11 is a sensor capable of measuring the amount of gas present in soil. The amount of gas present includes, for example, the gas concentration or the gas flow rate. In the present embodiment, the gas sensor 11 may be a sensor capable of measuring the gas concentration of any gas. For example, the gas sensor 11 may be a sensor capable of detecting the gas concentration of any gas selected from the group consisting of NO, N, NO, CH, and CO. The gas sensor 11 may be a sensor employing any of the following: an adsorption type, an NDIR type using light such as infrared, or a photoacoustic type using light such as infrared and sound. The adsorption type gas sensor 11 may be configured using a porous adsorbent material or a configuration mainly using metal oxide on a semiconductor.

[0020] Furthermore, the gas sensor 11 is not limited to a method for measuring gas concentration, and the objective of monitoring the amount of a desired gas can be achieved even if a method for measuring gas flow rate is used. Therefore, the gas sensor 11 may be a sensor using a thermal flow meter, an ultrasonic flow meter, a Coriolis flow meter (mass flow controller), or an ICP-MS.

[0021] The gas sensor 11 is placed in the soil in a farm field to measure the concentration of a gas in the soil surface layer or in the soil. For example, the gas sensor 11 may be placed in the soil surface layer. Alternatively, the gas sensor 11 may be placed so that at least a portion of the gas sensor 11 is buried in the soil. Generally, when the gas sensor 11 is buried in the soil, it is less susceptible to the influence of fluctuations in the gas fraction of atmospheric components than when the gas sensor 11 is placed in the soil surface layer, and the concentration of a desired gas can be measured with high sensitivity.

[0022] The detection device 10 transmits the gas concentration obtained by the gas sensor 11 to the control device 20 as analog data via a sensor amplifier as necessary.

[0023] The control device 20 includes an arithmetic circuit 21 , a storage device 22 , an input device 23 , an output device 24 , and a communication circuit 25 .

[0024] The arithmetic circuit 21 controls the overall operation of the control device 20. The arithmetic circuit 21 may be configured to realize a predetermined function through cooperation between hardware resources and software, or may be configured to realize a predetermined function using a dedicated hardware circuit.

[0025] As an example of the former, the arithmetic circuit 21 includes a general-purpose processor such as a CPU or MPU that executes a program to achieve predetermined processing or functions. The arithmetic circuit 21 is configured to be able to communicate with the storage device 22. The arithmetic circuit 21 reads and executes arithmetic programs, etc. stored in the storage device 22, thereby achieving various functions in the control device 20. As an example of the latter, the arithmetic circuit 21 includes an FPGA or an ASIC. As can be understood from the above, the arithmetic circuit 21 can be achieved using a semiconductor integrated circuit such as a CPU, MPU, GPU, FPGA, DSP, or ASIC.

[0026] The storage device 22 is a storage medium capable of storing various types of information. The information includes programs and data. For example, the storage device 22 stores calculation programs for implementing various functions according to the first embodiment. The storage device 22 also stores estimation models used in the processing according to the first embodiment. The storage device 22 is realized by, for example, a volatile or non-volatile semiconductor memory such as a DRAM, an SRAM, or a flash memory, an SSD, an HDD, or other storage devices, or an appropriate combination thereof.

[0027] The input device 23 functions as an input unit for inputting information from a user. The input device 23 includes one or more human-machine interface devices. The human-machine interface devices include, for example, a keyboard, a pointing device (such as a mouse or a trackball), a touchpad, and the like.

[0028] The output device 24 functions as an output unit for outputting information to a user. The output device 24 includes one or more human-machine interface devices. The human-machine interface devices include, for example, a display, a speaker, etc. The human-machine interface devices may also be input / output devices such as a display (e.g., a liquid crystal panel or an organic EL panel) equipped with an in-cell touch panel.

[0029] The communication circuit 25 is an interface device configured to connect to other devices or systems via a communication line, either wired or wirelessly. The interface device is capable of performing communication in accordance with wired communication standards such as USB (registered trademark) or Ethernet (registered trademark). The interface device is also capable of performing communication in accordance with wireless communication standards such as Wi-Fi (registered trademark), Bluetooth (registered trademark), and mobile phone lines.

[0030] For example, the arithmetic circuit 21 is configured to receive analog data regarding the gas concentration transmitted by the detection device 10 via the communication circuit 25. For example, the arithmetic circuit 21 may be configured to acquire digital data regarding the gas concentration by converting the received analog data via an AD conversion circuit. The arithmetic circuit 21 may acquire data via wireless communication from the detection device 10 placed in the soil, or may acquire data via wired communication from the detection device 10 stored in the placement device 30 after the detection device 10 is retrieved by the arm 32, which will be described in detail later.

[0031] The placement device 30 includes a drive device 31 capable of moving the sensing unit 2 and the placement device 30 in the field, and an arm 32 configured to place and retrieve the detection device 10 in and from the soil of the field. In the first embodiment, the placement device 30 is a wheeled robot. The placement device 30 is not limited to a wheeled robot, and any mobile device can be applied. The placement device 30 may be, for example, a drone. In the first embodiment, the drive device 31 is a wheel. The drive device 31 is not limited to a wheel, and any device capable of moving the placement device 30 can be applied. The drive device 31 may be, for example, a caterpillar or a propeller. The drive device 31 is an example of a mobile device.

[0032] The arm 32 is configured to be able to place the detection device 10 housed in the sensing unit 2 in the soil at or near the base of plants and crops in the field. The arm 32 is also configured to be able to retrieve the detection device 10 from the soil at or near the base of plants and crops in the field and to store it back in the sensing unit 2. The detection device 10 can be placed in and retrieved from the soil at or near the base of plants via the placement device 30 remotely or semi-automatically. The arm 32 is an example of a placement mechanism.

[0033] The management device 40 includes a cleaning device 41 and a calibration device 42. The cleaning device 41 functions as a cleaning unit that cleans the detection device 10 recovered from the soil. The cleaning device 41 is configured to clean, for example, by rinsing with water, and then dry the gas sensor 11 included in the detection device 10 that has been recovered, for example, to the sensing unit 2. In this way, the cleaning device 41 can remove soil adhering to the gas sensor 11 and remove acidic and alkaline components contained in the soil, thereby preventing the formation of rust.

[0034] The calibration device 42 functions as, for example, a calibration unit that calibrates the value of the gas sensor 11. For example, the calibration device 42 can calibrate the value of the gas sensor 11 in an atmospheric environment by controlling the temperature and pressure for desorbing gas adhering to the gas sensor 11. In this way, the calibration device 42 can ensure the measurement accuracy of the detection device 10.

[0035] [1-2. Operation] The operation of the arithmetic circuit 21 of the control device 20 according to the first embodiment will be outlined below. The arithmetic circuit 21 prepares in advance a relationship between gas concentration and microbial activity, and a relationship between microbial activity and the amount of microorganisms present that has increased due to the spraying of microorganisms. The arithmetic circuit 21 is then configured to estimate the microbial activity from the gas concentration using the prepared relationships and calculate the spray amount.

[0036] In the first embodiment, microbial activity refers to, for example, the number or abundance ratio of microorganisms primarily colonized on roots. For example, the number of microorganisms can be determined as the number of microorganisms colonized on roots of a predetermined size. For example, the abundance ratio of microorganisms can be determined as the ratio of the number of a predetermined microorganism among multiple microorganisms to the number of multiple microorganisms colonized on roots of a predetermined size. The number or abundance ratio of microorganisms can be measured, for example, by genetic testing or PCR testing.

[0037] First, the estimation model generation process by the arithmetic circuit 21 will be described. In the first embodiment, the arithmetic circuit 21 is configured to generate in advance an estimation model that learns the relationship between gas concentration and microbial activity, and to estimate microbial activity from a measured gas concentration using the estimation model. In the first embodiment, the estimation model generation process is executed by the arithmetic circuit 21, but the estimation model generation process may be executed by an arithmetic circuit other than the arithmetic circuit 21. The estimation model can be stored in the storage device 22.

[0038] The estimation model according to the first embodiment may be configured, for example, by a neural network having an input layer that receives microbial activity, an output layer that outputs gas concentrations, and one or more intermediate layers that extract features. The intermediate layers may include, for example, a convolutional layer, a pooling layer, and a fully connected layer. The estimation model is not limited to a neural network and may be configured using other machine learning techniques, such as a generative adversarial network.

[0039] 2 is a flowchart showing an example of an estimation model generation process executed by the arithmetic circuit 21. The arithmetic circuit 21 acquires, as microbial activity, the number or abundance ratio of microorganisms detected in soil, plants, and crops in a field or laboratory environment, mainly colonizing roots (S10). The arithmetic circuit 21 also acquires the gas concentration in the soil measured by the gas sensor 11 in association with the microbial activity of the soil (S12).

[0040] For example, the arithmetic circuit 21 may acquire, as the gas concentration, the average value of the numerical values ​​acquired from the gas sensor 11 within a certain period. The arithmetic circuit 21 may also acquire, as the gas concentration, the average value of a section that can be considered a saturated region among the numerical values ​​acquired from the gas sensor 11 within the certain period. The arithmetic circuit 21 may also acquire, as the gas concentration, the integrated value, maximum value, or minimum value of the values ​​acquired from the gas sensor 11 within the certain period. The arithmetic circuit 21 may acquire, as the gas concentration, the area value between the horizontal axis and the gas sensor value within the certain period in a graph generated with the horizontal axis representing time and the vertical axis representing the numerical values ​​acquired from the gas sensor 11. In this way, the arithmetic circuit 21 can handle the measurement signal from the sensor provided in the detection device 10 by compressing the dimension of the measurement signal.

[0041] The arithmetic circuit 21 acquires a plurality of combinations of microbial activity and gas concentration that are associated with each other. Then, the arithmetic circuit 21 uses the combinations as training data to train an estimation model (S14). The arithmetic circuit 21 uses training data in which microbial activity is input data and gas concentration is output data. The arithmetic circuit 21 acquires training data for each plant or crop and uses it for training, thereby generating an estimation model for each plant or crop. The estimation model may be further trained by edge learning using the data acquired by the arithmetic circuit 21.

[0042] Fig. 3 is a graph showing an example of the relationship between gas concentration and microbial activity in soil, as represented by an estimation model. In the graph shown in Fig. 3, the vertical axis represents gas concentration, and the horizontal axis represents microbial activity. The arithmetic circuit 21 acquires the relationship shown in Fig. 3 for each plant and crop. Therefore, in the first embodiment, the arithmetic circuit 21 can acquire the relationship between gas concentration and microbial activity for each plant and crop.

[0043] FIG. 4 is a flowchart showing an example of the state determination process executed by the arithmetic circuit 21.

[0044] In the first embodiment, the detection device 10 is placed in the soil at or near the base of a plant in a farm field by the placement device 30, and measures the gas concentration of the gas in the soil. When the detection device 10 measures the gas concentration and transmits it to the control device 20, the arithmetic circuit 21 of the control device 20 acquires the gas concentration from the gas sensor 11 (S20).

[0045] Next, the arithmetic circuit 21 estimates the microbial activity in the soil from the acquired gas concentration based on a previously prepared relationship between the gas concentration and the microbial activity (S22). In the first embodiment, the arithmetic circuit 21 is configured to input the acquired gas concentration into the estimation model generated by the estimation model generation process, and acquire the microbial activity by inverse analysis using the estimation model.

[0046] 3, when the calculation circuit 21 acquires the gas concentration EMy from the gas sensor 11, the calculation circuit 21 inputs the gas concentration EMy into the estimation model and acquires the current microbial activity AMx in the soil from which the gas concentration EMy was acquired. The calculation circuit 21 can determine the difference D1 between the microbial activity AMx and a predetermined target value AGx for the microbial activity.

[0047] After estimating the microbial activity AMx, the calculation circuit 21 calculates the amount of microorganisms to be sprayed on the soil using the estimated microbial activity AMx based on the relationship between the microbial activity prepared in advance and the amount of microorganisms increased by the spraying of the microorganisms (S24). In this specification, the amount of microorganisms increased by the spraying of the microorganisms is also simply referred to as the "increased amount of microorganisms."

[0048] The predetermined relationship between microbial activity and increased abundance can be obtained, for example, as follows. First, the arithmetic circuit 21 acquires microbial activity in the soil. After microorganisms are sprayed on the soil, the arithmetic circuit 21 acquires the amount of microorganisms in the soil increased by the spraying of the microorganisms and the microbial activity in the soil after the spraying of the microorganisms, in association with each other. After microorganisms are further sprayed on the soil, the arithmetic circuit 21 similarly acquires the increased abundance and the microbial activity, in association with each other. In this manner, the arithmetic circuit 21 acquires multiple combinations of the associated increased abundance and microbial activity for multiple increased abundances. Then, based on the multiple combinations, the arithmetic circuit 21 sets a function that approximately represents the relationship between the increased abundance and microbial activity, and can acquire the function as the predetermined relationship between microbial activity and increased abundance. The predetermined relationship between microbial activity and increased abundance can be stored in the storage device 22. In the first embodiment, the arithmetic circuit 21 determines the amount of microorganisms to be sprayed using the function.

[0049] In the first embodiment, the amount of microorganisms applied is assumed to be equivalent to the amount of microorganisms increased in the soil by applying the applied amount to the soil. The increased amount may be the difference between the amount of microorganisms in the soil before application and the amount of microorganisms in the soil after application.

[0050] FIG. 5 is a graph showing an example of the relationship between microbial activity in soil and increased abundance. In the graph shown in FIG. 5, the vertical axis represents increased abundance, and the horizontal axis represents microbial activity. The arithmetic circuit 21 according to the first embodiment acquires a function represented by the graph as a pre-prepared relationship between microbial activity and increased abundance. When the arithmetic circuit 21 acquires microbial activity for a given soil, it can use the pre-prepared function to calculate the increased abundance for that soil and calculate the desired application rate.

[0051] For example, the calculation circuit 21 can calculate the difference D2 between the increased abundance SMy in the soil and the target increased abundance SGy as the amount of microorganisms to be sprayed based on the difference D1 between the estimated microbial activity AMx and the target microbial activity AGx and the function. For example, the increased abundance SMy indicates the amount of microorganisms that has increased in the soil by previously spraying microorganisms. The calculation circuit 21 may calculate the increased abundance SMy corresponding to the estimated microbial activity AMx and the increased abundance SGy corresponding to the target microbial activity AGx based on the function, and calculate the difference D2 between the increased abundance SGy and the increased abundance SMy as the amount of microorganisms to be sprayed.

[0052] The optimal state of microbial activity may vary depending on the state of the field, the state of the plant, and the state of the crop. Therefore, the target value may be set arbitrarily for each field. For example, a user may select several well-grown plants from a desired plant in a desired field, measure the number and ratio of microorganisms that are primarily colonized in the roots, calculate an average value from the measured values, and set the average value as the target value.

[0053] The method for selecting individuals in good condition may be judgment by human eyes, or judgment using AI or deep learning. For example, by taking images of plants and crops and accumulating image data of individuals in good condition, and using that data to train AI, individuals in good condition can be judged by AI.

[0054] The arithmetic circuit 21 may be configured to generate an estimation model in advance that has learned the relationship between microbial activity in soil and the increased abundance in the soil, and to input the microbial activity AMx and the target microbial activity AGx into the estimation model to calculate the application amount. For example, the arithmetic circuit 21 can generate training data using the microbial activity in a specific soil as input data and the increased abundance in the soil as output data, and train the estimation model using training data for multiple soils to generate the estimation model.

[0055] Next, an example of a method for operating the condition determination system 1 in a field and mapping the amount of microorganisms to be sprayed in the field and displaying it on the output device 24 will be described. In the first embodiment, the arithmetic circuit 21 calculates the amount of microorganisms to be sprayed on a specific soil in the field, and then displays the amount of spraying on the output device 24.

[0056] When the gas concentration is transmitted from the detection device 10 to the control device 20, the condition determination system 1 according to the first embodiment places the detection device 10 in soil at a different position in the field. For example, the condition determination system 1 drives the arm 32 of the placement device 30 to place the detection device 10 in soil at a position where the gas concentration has not been measured. The condition determination system 1 may drive the drive device 31 to move to a position in the field that is different from the current position, and place the detection device 10 using the arm 32.

[0057] For example, in order to execute the condition determination process by the control device 20, the condition determination system 1 moves the placement device 30 on relatively flat land between ridges and places the detection device 10 in the soil of each plant or crop using the arm 32. The gas sensor 11 of the detection device 10 measures gases in the soil or in the surface layer of the soil.

[0058] Because the amount of gas released from soil is minute, for example, the gas sensor 11 acquires data in the range of 0.1 to 50.0 ppm in a single measurement, taking approximately 1 to 10 minutes. In the first embodiment, the sensing unit 2 includes two detection devices 10. While the arithmetic circuit 21 measures the gas concentration using one detection device 10, the arithmetic circuit 21 cleans and dries each sensor of the other detection device 10 using the cleaning device 41 and calibrates each sensor using the calibration device 42. This allows the sensing unit 2 to suppress errors due to cumulative measurements. The sensing unit 2 may include one, or three or more detection devices 10.

[0059] If a single measurement takes approximately 1 to 10 minutes, gas concentration measurements may be performed over several days depending on the size of the field in order to obtain measurement results for the entire field. Therefore, in the condition determination system 1 according to the first embodiment, the placement device 30 is configured to be able to learn the travel route of the drive device 31 using, for example, a map or image of the field. The placement device 30 is also configured to be able to learn how to place and retrieve the detection device 10 in the soil using the arm 32. By learning the travel route, the placement device 30 can travel autonomously day or night using the drive device 31. Furthermore, by learning how to place and retrieve the detection device 10, the placement device 30 can automatically place and retrieve the detection device 10 using the arm 32. Therefore, the condition determination system 1 can automatically acquire measurement data day or night.

[0060] FIG. 6 is a diagram showing an example of a heat map showing estimated microbial activity in a field, displayed on the output device 24 of the control device 20 according to the first embodiment. The output device 24 may also display a coefficient of variation for the distribution of microbial activity together with the heat map. In the example shown in FIG. 6, the coefficient of variation is 29.5%. In the heat map shown in FIG. 6, the field is divided into a plurality of regions in a matrix. For example, the arithmetic circuit 21 acquires gas concentrations for the soil in each region and estimates the microbial activity in that region. The shades of color in the heat map represent the relative magnitude of the estimated microbial activity in each region, i.e., the current value of the microbial activity. In the heat map, the estimated microbial activity in a darker region is greater than the estimated microbial activity in a lighter region.

[0061] Referring to the heat map shown in FIG. 6, it can be seen that the area located at the center of the field has a higher microbial activity than the areas located around the center. It can also be seen that the lower area of ​​the field has a higher microbial activity than the upper area. Because the field is located above a waterfront, it can be seen that the distribution between the upper and lower areas of the field is affected by the surrounding environment. In the condition determination system 1 according to the present disclosure, the arithmetic circuit 21 determines the amount of microbial application to suppress such differences in the distribution of current microbial activity values. The arithmetic circuit 21 can also map the application amount in the field and display it on the output device 24.

[0062] 7 is a diagram showing an example of a heat map that maps the amount of microorganisms applied to a farm field, and is displayed on the output device 24 of the control device 20 according to the first embodiment. In the heat map shown in FIG. 7, the farm field is divided into a plurality of regions in a matrix. The shades of color in the heat map indicate the relative magnitude of the application amount calculated for each region. In the heat map, the application amount calculated for the darker colored regions is greater than the application amount calculated for the lighter colored regions.

[0063] 7, it can be seen that the spray rate for the area located at the center of the field is smaller than the spray rate for the areas located around the center area, and that the spray rate for the lower area of ​​the field is larger than the spray rate for the upper area.

[0064] The condition determination system 1 according to the first embodiment can map the entire farm field and visualize the differences in microbial activity and application amount, thereby enabling plants and crops to be harvested uniformly at a high level.

[0065] The microbial flora used for training the estimation model in the state determination system 1 according to the first embodiment may include various microbial species. For example, the microbial species include bacteria, actinomycetes, filamentous fungi, mycorrhizal fungi, algae, nitrifying bacteria, rhizobia, denitrifying bacteria, methanogens, methanogenic archaea, and cyanobacteria. Microorganisms generally decompose organic matter in soil to produce CO2, and are therefore considered to have a strong relationship with CO2 emissions.

[0066] Nitrifying bacteria, rhizobia, denitrifying bacteria, and cyanobacteria play important roles in the process of absorbing nitrogen from the atmosphere and soil, such as producing nitrate ion-based substances, reducing them to N2O and N2, and fixing nitrogen gas. They are thought to be particularly closely related to the release of N2O and N2. For example, nitrifying bacteria include ammonia nitrifiers and nitrite oxidizers. For example, nitrifying bacteria produce nitrate nitrogen from ammonia nitrogen. For example, rhizobia fix atmospheric nitrogen gas. For example, denitrifying bacteria produce nitrogen gas from nitrite nitrogen and release it into the atmosphere as a gas.

[0067] Furthermore, methanogens and methanogenic archaea play a role in generating CO2 in fields where oxygen is readily available, using acetic acid or hydrogen as a base during the process of decomposing organic matter and organic carbon (C) in the soil. Furthermore, methanogens and methanogenic archaea play a role in generating CH4 in fields where oxygen is poorly supplied, such as rice paddies. Therefore, methanogens and methanogenic archaea are thought to have a particularly strong relationship with the release of CH4 or CO2.

[0068] Additionally, filamentous fungi and mycorrhizal fungi are known to exist as species that have a strong influence on certain plants and crops. They fix nitrogen in the process of absorbing nitrogen from the atmosphere and soil, and mineralize nitrogen in the process of decomposing organic matter and organic carbon (C) in the soil. Therefore, filamentous fungi and mycorrhizal fungi are thought to have a strong relationship, particularly with the emission of N2O, N2, CH4, and CO2.

[0069] The gas species used for training the estimation model in the state determination system 1 according to the first embodiment may include various gas species. N2O, N2, NO, CH4, or CO2 is considered to have a strong relationship with microbial activity in the process of nitrogen uptake or the process of decomposing organic matter and organic carbon. Therefore, the gas species may include, for example, any one selected from the group consisting of N2O, N2, NO, CH4, or CO2.

[0070] CO2 may exist in the atmosphere at a predetermined rate. Therefore, the arithmetic circuit 21 may be configured to subtract a predetermined value from the acquired gas concentration of CO2 in order to obtain the concentration of CO2 emitted by microorganisms. In this manner, the arithmetic circuit 21 may be configured to perform some kind of correction process on the acquired gas concentration for each gas type.

[0071] In the state determination system 1 according to the first embodiment, the gas sensor 11 measures the concentration of a gas released from a microorganism on the soil surface or in the soil. For example, the gas sensor 11 measures a gas collected by the gas sensor 11 on the soil surface or in the soil. Similarly, in the estimation model generation process, the arithmetic circuit 21 can generate an estimation model using the gas concentration of the gas collected by the gas sensor 11 on the soil surface or in the soil.

[0072] The microorganisms colonizing the roots may include multiple microorganisms. For example, if the total number of microorganisms colonizing the roots is Z, Z includes the number A of microorganisms a, the number B of microorganisms b, and the number C of microorganisms c. As described above, a given microorganism may have a strong relationship with the emission of a given gas. For example, microorganisms a and b may be related to the emission of NO, and microorganisms b and c may be related to the emission of CO. In this case, the NO gas concentration E N2O can be estimated as a function of the number A of microorganisms a and the number B of microorganisms b, as shown in Equation (1). CO2 can be estimated as a function of the number B of microorganisms b and the number C of microorganisms c, as shown in equation (2). Equation (1) or (2) is an example of the relationship between gas concentration and microbial activity shown in Figure 3. E N2O =f1(A+B) (1) E CO2 =f2(B+C) (2)

[0073] When microorganisms are measured by PCR testing, a portion of the total number Z of microorganisms described above may be detected. For example, the measurement value Y of a microorganism detected by PCR testing is expressed as Y = α × Z, where α is a predetermined coefficient. Also, for example, the measurement value V of microorganism a detected by PCR testing is expressed as V = β × A. The measurement value W of microorganism b detected by PCR testing is expressed as W = β × B. The measurement value X of microorganism c detected by PCR testing is expressed as X = β × C, where β is a predetermined coefficient.

[0074] When microbial activity is defined as the number of microorganisms present, Figure 3 shows the gas concentration EN2O When showing the graph for gas concentration E N2O Regarding the gas concentration E, the current microbial activity AMx on the horizontal axis shown in FIG. 3 is expressed by the sum of V and W. In other words, the microbial activity AMx is expressed as AMx=β×(A+B). Similarly, CO2 The current microbial activity of B can be expressed as the sum of the above-mentioned W and X, that is, "β×(B+C)".

[0075] As described above, the calculation circuit 21 can set a function that approximately represents the relationship between the increased abundance and the microbial activity. For example, the increased abundance SM of microorganisms present in the soil with respect to N2O can be expressed as N2O and gas concentration E N2O The function of the current microbial activity AMx for N2O can be expressed as Equation (3). N2O and gas concentration E N2O The function of the target microbial activity AGx for the amount of increased abundance SM can be expressed by equation (4). Here, η1 is a predetermined coefficient. A0 is the microbial activity when the amount of increased abundance SM is zero. N2O and increased abundance SG N2O The difference between the spray amount D N2O Similarly, the increased abundance of microorganisms present in the soil with respect to CO2, SM CO2 , and the target increased abundance SG CO2 can also be expressed by a predetermined formula. SM N2O =η1×(AMx-A0) (3) SG N2O =η1×(AGx-A0) (4)

[0076] The calculation circuit 21 determines the ratio of the spray amount Da of the microorganism a, the spray amount Db of the microorganism b, and the spray amount Dc of the microorganism c so that it is equal to the ratio of the numbers A to C of the microorganisms a to c. N2O is the sum of the spray amount Da and the spray amount Db. Therefore, the calculation circuit 21 calculates the ratio of the numbers A to C of the microorganisms and the spray amount D N2O Based on this, the application amount Da, the application amount Db, and the application amount Dc can be calculated.

[0077] When microbial activity is defined as the ratio of microorganisms present, Figure 3 shows the gas concentration E N2O When showing the graph for gas concentration E N2O Regarding the gas concentration E, the current microbial activity AMx on the horizontal axis shown in FIG. 3 is expressed as AMx = β × ((A + B) / Y). CO2 The microbial activity of the above is expressed as "β x ((B + C) / Y)".

[0078] For example, the increased abundance of microorganisms present in soil in relation to NO N2O and gas concentration E N2O The function of the current microbial activity AMx for N2O can be expressed as Equation (5). N2O and gas concentration E N2O The function of the target microbial activity AGx for the amount of microbial activity can be expressed by Equation (6), where η2 is a predetermined coefficient. N2O and increased abundance SG N2O The difference between the spray amount D N2O Similarly, the increased abundance of microorganisms present in the soil with respect to CO2, SM CO2 , and the target increased abundance SG CO2 can also be expressed by a predetermined formula. SM N2O =η2×(AMx-A0) (5) SG N2O =η2×(AGx-A0) (6)

[0079] As in the case where microbial activity is determined as the number of microorganisms present, the arithmetic circuit 21 can calculate the spraying amounts Da, Db, and Dc. For example, the arithmetic circuit 21 can determine the spraying amount Da so that it is proportional to the abundance ratio A / Y of microorganism a, the spraying amount Db so that it is proportional to the abundance ratio B / Y of microorganism b, and the spraying amount Dc so that it is proportional to the abundance ratio C / Y of microorganism c. Even when the abundance ratio of microorganisms is used as the microbial activity in this way, the target abundance of microorganisms in the soil can be ensured by, for example, a user spraying the amount calculated by the arithmetic circuit 21 onto the soil.

[0080] In the present disclosure, a farm field refers to a place where agricultural crops are cultivated. Farm fields include, for example, rice paddies, greenhouses, orchards, greenhouses, forests, fallow land, and the like.

[0081] [1-3.Effects] According to the state determination method according to the first embodiment of the present disclosure, the following effects can be achieved.

[0082] The method for determining the state of a farm field executed by the arithmetic circuit 21 includes a step of acquiring the amount of gas present in the surface soil of the farm field or in the soil, and a step of estimating the microbial activity in the soil from the acquired amount of gas present, based on a previously prepared relationship between the amount of gas present and the microbial activity.

[0083] According to this method, the arithmetic circuit 21 can acquire the amount of gas present in the soil, such as the gas concentration, and estimate the microbial activity in the soil from the amount of gas present. If the microbial activity can be estimated, the state of the microorganisms in the soil that contribute to the promotion of plant and crop growth can be inferred, and the arithmetic circuit 21 can appropriately estimate the state of the field. Since the user of the state determination method can grasp the state of the field, processing can be performed according to the state of the field.

[0084] The condition determination method further includes a step of calculating the amount of microorganisms to be sprayed on the soil from the estimated microbial activity. According to this method, the arithmetic circuit 21 can calculate the amount of microorganisms to be sprayed on the soil based on the estimated microbial activity of the soil. Since the user of the condition determination method can determine the amount of microorganisms to be sprayed on the soil, the user can efficiently spray microorganisms on the soil, which can effectively promote the growth of plants and crops.

[0085] The condition determination method further includes a step of outputting a map showing the spray amount at multiple points in the field. This method allows the arithmetic circuit 21 to visualize the spray amount for each position in the field and also visualize the difference in spray amount depending on the position. Because the user can understand the appropriate spray amount for the soil in relation to the position in the field, the microorganisms can be sprayed efficiently, which can effectively promote the growth of plants and crops.

[0086] In the condition determination method, the gas includes any one selected from the group consisting of NO, N, NO, CH, and CO. In this way, the arithmetic circuit 21 can estimate the microbial activity of the soil based on the amount of any one gas selected from the above group. Therefore, a user of the condition determination method can grasp the condition of the soil by estimating the microbial activity of the soil using a gas suitable for the microorganisms used.

[0087] In the state determination method, the estimating step includes estimating the microbial activity in the soil from the acquired gas using machine learning or multivariate analysis. In this way, the arithmetic circuit 21 can estimate the microbial activity in the soil using various methods.

[0088] In the condition determination method, the acquiring step includes acquiring gas abundances at multiple locations in the field in association with the positions in the field. The estimating step includes estimating microbial activity at the multiple locations from the multiple acquired gas abundances based on a predetermined relationship between the gas abundance and microbial activity. The calculating step includes calculating the microbial application rates at the multiple locations from the multiple estimated microbial activity rates. The condition determination method further includes a step of mapping the calculated application rates in association with the positions in the field. According to this method, the arithmetic circuit 21 can visualize the application rate for each position in the field and can also visualize differences in application rate depending on the position. Since the user can understand the appropriate application rate for the soil in association with the position in the field, microorganisms can be sprayed efficiently, which can effectively promote the growth of plants and crops.

[0089] The condition determination device 20 includes a calculation circuit 21 and a storage device 22 that stores the relationship between the amount of gas present and microbial activity. The calculation circuit 21 is configured to acquire the amount of gas present in the soil measured by the sensor 11 in the surface soil or in the soil based on the relationship stored in the storage device 22, and estimate the microbial activity in the soil from the acquired amount of gas present. With this configuration, the calculation circuit 21 can acquire the amount of gas present in the soil and estimate the microbial activity in the soil from the amount of gas present. Estimating the microbial activity allows the state of microorganisms in the soil that contribute to the growth of plants and crops to be inferred, and the calculation circuit 21 can appropriately estimate the condition of the field. This allows the user to perform processing according to the condition of the field.

[0090] The condition determination system 1 includes a sensor 11, a placement device 30, the above-described condition determination device 20, and a sensor management device 40. The sensor 11 measures the amount of gas present in the surface soil and in the soil in a field. The placement device 30 includes a placement mechanism 32 capable of placing the sensor 11 in the soil and a moving device 31 capable of moving the sensor 11 around the field. The sensor management device 40 includes a cleaning device 41 capable of removing dirt adhering to the sensor 11 and a calibration device capable of calibrating the sensor 11. With this configuration, the condition determination system 1 can move around the field automatically, partially automatically, or via remote control by a user, and estimate microbial activity at multiple locations in the field. Because microbial activity can be estimated at multiple locations in the field, the user can grasp the overall condition of the field and perform appropriate processing for each location in the field.

[0091] The estimation model training method executed by the arithmetic circuit 21 includes the steps of acquiring the abundance of gases measured in the surface layer of soil and in the soil, and acquiring the microbial activity of the soil. The training method also includes the step of training an estimation model that estimates microbial activity from the abundance of gases by performing training using training data that includes the acquired microbial activity as input data and the acquired amount of gas present as output data. According to this method, the arithmetic circuit 21 can generate an estimation model for estimating microbial activity of soil from the gases measured in the soil.

[0092] (Embodiment 2) [2-1.Configuration] An overview of the condition determination system 1 according to embodiment 2 will be described. Fig. 8 shows a schematic diagram of the condition determination system 1 according to embodiment 2. The detection device 10 of the condition determination system 1 according to embodiment 2 further includes a temperature sensor 12, as compared to the detection device 10 of the condition determination system 1 according to embodiment 1.

[0093] The temperature sensor 12 measures the temperature in the measurement environment for the gas concentration measured by the gas sensor 11 and transmits the measured temperature to the control device 20. This allows the arithmetic circuit 21 of the control device 20 to use the soil temperature as a parameter for estimating microbial activity. Temperature is an example of soil environment information. The soil environment information indicates various types of information about the soil in which each sensor of the detection device 10 is placed. For example, the soil environment information may be measurement conditions including the state of the soil when measurement was performed by each sensor of the detection device 10.

[0094] [2-2. Operation] Below, an outline of the operation of the arithmetic circuit 21 of the control device 20 according to embodiment 2 will be described. In embodiment 2, the arithmetic circuit 21 is configured to prepare in advance relationships between gas concentration, temperature, and microbial activity, and to estimate microbial activity from the gas concentration and temperature using the prepared relationships.

[0095] Specifically, the arithmetic circuit 21 is configured to generate in advance an estimation model that learns the relationship between gas concentration, temperature, and microbial activity, and to use the estimation model to estimate microbial activity from the measured gas concentration and temperature.

[0096] 9 is a flowchart showing an example of an estimation model generation process executed by the arithmetic circuit 21 in the second embodiment. The arithmetic circuit 21 acquires, as microbial activity, the number or abundance ratio of microorganisms detected in soil, plants, and crops in a field or laboratory environment, which are mainly colonized in roots (S30). The arithmetic circuit 21 acquires soil environment information at the time of acquiring the microbial activity in association with the microbial activity (S32). Specifically, the arithmetic circuit 21 acquires the temperature at the time of acquiring the microbial activity, measured by the temperature sensor 12. The arithmetic circuit 21 also acquires the gas concentration in the soil measured by the gas sensor 11 in association with the microbial activity and temperature of the soil (S34).

[0097] The arithmetic circuit 21 acquires a plurality of combinations of microbial activity, temperature, and gas concentration that are associated with each other. Then, the arithmetic circuit 21 uses the combinations as training data to train an estimation model (S36). The arithmetic circuit 21 uses the training data in which the microbial activity and temperature are input data and the gas concentration is output data.

[0098] Fig. 10 is a graph showing an example of the relationship between gas concentration, temperature, and microbial activity in soil, as represented by an estimation model. In the graph shown in Fig. 10, the vertical axis represents gas concentration, and the horizontal axis represents microbial activity. In the graph shown in Fig. 10, line T1 represents the relationship between gas concentration and microbial activity when the temperature is 22°C. Line T2 represents the relationship between gas concentration and microbial activity when the temperature is 30°C.

[0099] Generally, the higher the temperature, the more active the decomposition reaction of nutrients such as nitrogen and organic matter caused by the activity of the microbial flora. In this way, the activity of the microbial flora can be affected by environmental conditions. The state determination system 1 according to the second embodiment can include temperature as a parameter of the estimation model. Therefore, the state determination system 1 can reflect the relationship due to differences in measurement environmental conditions in the estimation model, and can estimate the microbial activity level with higher accuracy than the state determination system 1 according to the first embodiment.

[0100] As shown in FIG. 10, the gas concentration EMy acquired from the gas sensor 11 corresponds to microbial activity, which varies depending on the temperature. For example, when the temperature acquired from the temperature sensor 12 is T1, the arithmetic circuit 21 uses the estimation model to estimate microbial activity AMx from the gas concentration EMy and temperature T1. When the temperature acquired from the temperature sensor 12 is T2, the arithmetic circuit 21 uses the estimation model to estimate microbial activity AMx2 from the gas concentration EMy and temperature T2. In this way, the relationship between gas concentration and microbial activity can change depending on the temperature. The arithmetic circuit 21 according to the second embodiment can improve the accuracy of estimating microbial activity by using temperature in addition to gas concentration.

[0101] FIG. 11 is a flowchart showing an example of the state determination process executed by the arithmetic circuit 21 in the second embodiment.

[0102] In the second embodiment, the detection device 10 is placed in the soil at or near the base of a plant in a farm field by the placement device 30, and measures the temperature as well as the gas concentration in the soil. When the detection device 10 measures the gas concentration and temperature and transmits them to the control device 20, the arithmetic circuit 21 of the control device 20 acquires the gas concentration measured by the gas sensor 11 (S40) and the temperature measured by the temperature sensor 12 (S42).

[0103] The arithmetic circuit 21 estimates the microbial activity in the soil from the acquired gas concentration and temperature based on a prepared relationship between the gas concentration, temperature, and microbial activity (S44). In the second embodiment, the arithmetic circuit 21 is configured to input the acquired gas concentration and temperature into the generated estimation model and acquire the microbial activity.

[0104] 10, the arithmetic circuit 21 acquires the gas concentration EMy from the gas sensor 11 and the temperature T2 from the temperature sensor 12, and then inputs the gas concentration EMy and the temperature T2 into the estimation model to acquire the microbial activity AMx2. The arithmetic circuit 21 can determine the difference D3 between the microbial activity AMx2 and a predetermined target value AGx for the microbial activity.

[0105] After estimating the microbial activity, the calculation circuit 21 calculates the amount of microorganisms to be spread on the soil from the estimated microbial activity based on a pre-prepared relationship between the microbial activity and the increased abundance (S46).

[0106] Figure 12 is a graph showing an example of the relationship between microbial activity in soil and increased abundance, where the vertical axis represents increased abundance and the horizontal axis represents microbial activity.

[0107] For example, the calculation circuit 21 can calculate the difference D4 between the increased abundance SMy2 in the soil and the target increased abundance SGy as the amount of microorganisms to be sprayed based on the difference D3 between the estimated microbial activity AMx2 and the target microbial activity AGx and the function.

[0108] Just as the relationship between gas concentration and microbial activity shown in Figure 10 changes with temperature, it is expected that the relationship between microbial activity and increased abundance shown in Figure 12 will also change. Compared to the changes in the graph shown in Figure 10, the changes in the relationship between microbial activity and increased abundance are so small that they can be ignored in practice. The relationship between microbial activity and increased abundance changes so that, for example, when the temperature is high, the slope of the graph shown in Figure 12 becomes smaller. In other words, as the temperature increases, microbial activity becomes activated with a smaller amount of microorganisms sprayed.

[0109] In this way, the condition determination system 1 according to the present disclosure can estimate microbial activity using soil environmental information in addition to gas concentration, thereby improving estimation accuracy compared to estimating microbial activity using only gas concentration.

[0110] [2-3. Effects] According to the state determination method according to the second embodiment of the present disclosure, the following effects can be achieved.

[0111] The field condition determination method executed by the arithmetic circuit 21 includes a step of acquiring the amount of gas present in the surface soil of the field or in the soil, measured. The condition determination method further includes a step of estimating the microbial activity in the soil from the acquired amount of gas present, based on a previously prepared relationship between the amount of gas present and the microbial activity. The condition determination method further includes a step of acquiring soil environmental information about the soil. The estimating step includes estimating the microbial activity in the soil from the acquired amount of gas present and the acquired soil environmental information, based on a previously prepared relationship between the amount of gas present, the soil environmental information, and the microbial activity.

[0112] According to this method, the arithmetic circuit 21 can estimate the microbial activity in the soil using the soil environmental information in addition to the amount of gas present. The release of gas due to microbial activity can be more active at higher temperatures. For example, by using temperature as soil environmental information, the arithmetic circuit 21 can estimate the microbial activity with higher accuracy than when temperature is not used.

[0113] (Embodiment 3) [3-1.Configuration] An overview of a condition determination system 1 according to embodiment 3 will be described. Fig. 13 shows a schematic diagram of the condition determination system 1 according to embodiment 3. Compared to the detection device 10 of the condition determination system 1 according to embodiment 2, the detection device 10 of the condition determination system 1 according to embodiment 3 further includes a moisture sensor 13, a pH sensor 14, and an oxygen concentration sensor 15.

[0114] The moisture sensor 13 measures the moisture amount in the environment whose gas concentration has been acquired by the gas sensor 11, and transmits the moisture amount to the control device 20. The pH sensor 14 measures the pH value of the environment and transmits the pH value to the control device 20. The oxygen concentration sensor 15 measures the oxygen concentration in the environment and transmits the oxygen concentration to the control device 20. This allows the arithmetic circuit 21 of the control device 20 to use the moisture amount, pH value, and oxygen concentration of the soil as parameters for estimating microbial activity. The moisture amount, pH value, and oxygen concentration are examples of soil environment information.

[0115] In order to increase the measurement accuracy, the detection device 10 may be provided with a moisture sensor 13, a pH sensor 14, etc., individually. In addition, in order to prevent an increase in the system size, the detection device 10 may be provided with a composite sensor that combines the functions of the temperature sensor 12, the moisture sensor 13, and the pH sensor 14 into one sensor.

[0116] [3-2. Operation] An outline of the operation of the arithmetic circuit 21 of the control device 20 according to the third embodiment will be described below. In the third embodiment, the arithmetic circuit 21 is configured to prepare relationships between the gas concentration, temperature, moisture content, pH value, and oxygen concentration and the microbial activity in advance. The arithmetic circuit 21 is then configured to estimate the microbial activity from the gas concentration, temperature, moisture content, pH value, and oxygen concentration using the prepared relationships.

[0117] As in the first and second embodiments, the arithmetic circuit 21 is configured to generate an estimation model that has previously learned the relationships between the gas concentration, temperature, moisture content, pH value, and oxygen concentration and the microbial activity. The arithmetic circuit 21 is also configured to estimate the microbial activity from the acquired gas concentration, temperature, moisture content, pH value, and oxygen concentration using the estimation model.

[0118] The arithmetic circuit 21 learns the estimation model in the same manner as the processing described in the second embodiment. Specifically, the arithmetic circuit 21 acquires a plurality of combinations of microbial activity, temperature, moisture content, pH value, oxygen concentration, and gas concentration, which are associated with each other. The arithmetic circuit 21 then learns the estimation model using the combinations. The arithmetic circuit 21 uses training data in which the microbial activity, temperature, moisture content, pH value, and oxygen concentration are input data and the gas concentration is output data.

[0119] The calculation circuit 21 acquires gas concentration, temperature, moisture content, pH value, and oxygen concentration via the detector 10 placed in the soil of the field, and estimates the current microbial activity in the soil using the acquired data and an estimation model.The calculation circuit 21 then calculates the application amount using a previously prepared relationship between microbial activity and increased abundance.

[0120] The state determination system 1 according to the third embodiment can include temperature, moisture content, pH value, and oxygen concentration as parameters of the estimation model. Therefore, the state determination system 1 can reflect the relationship due to differences in environmental conditions in the estimation model, and can estimate microbial activity with higher accuracy than the state determination system 1 according to the second embodiment.

[0121] [3-3. Effects] The state determination method according to the third embodiment of the present disclosure can provide the following effects.

[0122] The field condition determination method executed by the arithmetic circuit 21 includes a step of acquiring the amount of gas present in the surface layer of the soil or in the soil measured in the field. The condition determination method further includes a step of estimating the microbial activity in the soil from the acquired amount of gas present, based on a previously prepared relationship between the amount of gas present and the microbial activity. The condition determination method further includes a step of acquiring soil environmental information for the soil. The estimating step includes estimating the microbial activity in the soil from the acquired amount of gas present and the acquired soil environmental information, based on a previously prepared relationship between the amount of gas present, the soil environmental information, and the microbial activity. The soil environmental information for the soil includes any one selected from the group consisting of temperature, moisture content, oxygen concentration in the surface layer of the soil or in the soil, and pH value.

[0123] According to this method, the arithmetic circuit 21 can estimate the microbial activity in the soil using soil environmental information in addition to the amount of gas present. The gas released by microbial activity can vary depending on the soil environment. By using information such as temperature and moisture content as soil environmental information, the arithmetic circuit 21 can estimate the microbial activity with higher accuracy than when not using soil environmental information.

[0124] (Variation) In the condition determination system 1 according to the present disclosure, the arithmetic circuit 21 creates an estimation model and uses the estimation model to estimate microbial activity from soil-related data such as gas concentrations, but the method for estimating microbial activity is not limited to using an estimation model. For example, the arithmetic circuit 21 may be configured to acquire in advance a relationship between soil-related data and microbial activity, and estimate microbial activity from soil-related data based on the relationship using multivariate analysis.

[0125] The condition determination system 1 according to the present disclosure primarily uses the number or ratio of microorganisms colonizing roots as the microbial activity level, but the microbial activity level is not limited to the number or ratio of microorganisms colonizing roots. For example, certain microorganisms known as endophytes or plant symbiotic bacteria have a strong influence on plants and crops, and are believed to exert their influence by colonizing not only roots but also the interior of the stems of plants and crops. Therefore, the condition determination system 1 may measure the number or ratio of microorganisms colonizing the interior of plants and crops, such as stems, using a genetic test or PCR test, and use this as the microbial activity level. Furthermore, the condition determination system 1 may use the number or ratio of microorganisms measured by collecting soil around the roots, rather than the number or ratio of microorganisms colonizing roots, as the microbial activity level.

[0126] (Summary of aspects) As is clear from the above description, the present disclosure includes the following aspects. In the following, reference numerals are given in parentheses only to clarify the correspondence with the embodiments.

[0127] (Aspect 1) A method for determining a state of a field according to the present disclosure is a method for determining a state of a field executed by an arithmetic circuit (21), A step of acquiring the amount of gas present in the surface soil or the soil in the field; a step of estimating microbial activity in the soil from the acquired gas abundance based on a previously prepared relationship between the gas abundance and microbial activity; Includes:

[0128] (Aspect 2) The state determination method of aspect 1 may further include a step of calculating the amount of microorganisms to be sprayed on the soil from the estimated microbial activity.

[0129] (Aspect 3) The condition determination method of aspect 2 may further include the step of outputting a map showing the application amounts at a plurality of points in the field.

[0130] (Aspect 4) The state determination method of any one of Aspects 1 to 3 further includes a step of acquiring soil environment information of the soil, The estimating step may include estimating the microbial activity in the soil from the acquired gas abundance and the acquired soil environmental information based on a pre-prepared relationship between the gas abundance, soil environmental information, and microbial activity.

[0131] (Aspect 5) In the condition determination method of Aspect 4, the soil environment information of the soil may include any one selected from the group consisting of temperature, moisture content, oxygen concentration in the soil surface or soil, and pH value.

[0132] (Aspect 6) In the state determination method of any one of Aspects 1 to 5, the gas may include any one selected from the group consisting of N2O, N2, NO, CH4, and CO2.

[0133] (Aspect 7) In any of the state determination methods of Aspects 1 to 6, the estimating step may include estimating the microbial activity in the soil from the acquired gas using machine learning or multivariate analysis.

[0134] (Aspect 8) In the state determination method of Aspect 2, the acquiring step includes acquiring the gas abundance of the gas at a plurality of locations in the field in association with the location in the field; the estimating step includes estimating the microbial activity at the plurality of locations from the plurality of acquired amounts of gas present based on a relationship between the amount of gas present and the microbial activity prepared in advance; the calculating step includes calculating the amount of microorganisms dispersed at the plurality of locations from the plurality of estimated degrees of microbial activity; The condition determination method may further include a step of mapping the calculated application amount in association with a position in the field.

[0135] (Aspect 9) A state determination device (20) according to the present disclosure includes an arithmetic circuit (21) and a storage device (22) that stores a relationship between the amount of gas present and the microbial activity, The arithmetic circuit comprises: Acquiring, from a sensor (11) placed in soil in a field, the amount of gas present in the surface soil or in the soil measured by the sensor; estimating a microbial activity in the soil from the acquired gas abundance based on the relationship stored in the storage device; is configured to execute

[0136] (Aspect 10) The condition determination system according to the present disclosure includes a sensor (11) for measuring the amount of gas present in the surface soil or the soil in a farm field; a placement device (30) including a placement mechanism (32) capable of placing the sensor in the soil and a moving device (31) capable of moving in the field; A state determination device (20) of aspect 9; a sensor management device (40) including a cleaning device (41) capable of removing dirt adhering to the sensor and a calibration device (42) capable of calibrating the sensor; Equipped with.

[0137] (Aspect 11) A learning method according to the present disclosure is a learning method for an estimation model executed by an arithmetic circuit (21), obtaining gas abundances of gases measured in and over the soil surface; Obtaining the microbial activity of the soil; a step of learning the estimation model that estimates the microbial activity from the amount of gas present by performing learning using training data including the acquired microbial activity as input data and the acquired amount of gas present as output data; Includes:

[0138] As used herein, terms such as "first," "second," etc. are used for descriptive purposes only and should not be understood as expressing or implying the relative importance or ranking of technical features. Features qualified as "first" and "second" expressly or imply the inclusion of one or more of that feature.

[0139] The state determination method, state determination device, state determination system, and learning method described in the present disclosure are realized by cooperation between hardware resources, such as a processor and memory, and software (computer program). [Industrial Applicability]

[0140] According to the present disclosure, a state determination method, a state determination device, a state determination system, and a learning method that can estimate the state of a farm field can be provided, and therefore can be suitably used in this type of industrial field. [Explanation of symbols]

[0141] 1. Status determination system 2 Sensing Unit 10. Detection Device 11 Gas Sensor 12 Temperature Sensor 20 Control device 21 Arithmetic circuit 22 Storage device 24 Output Devices 30 Placement device 31 Drive unit 32 Arm 40 Management device 41 Cleaning equipment 42 Calibration device

Claims

1. A method for determining a state of a farm field executed by an arithmetic circuit, comprising: A step of acquiring the amount of gas present in the surface soil or the soil in the field; a step of estimating microbial activity in the soil from the acquired gas abundance based on a previously prepared relationship between the gas abundance and microbial activity; A method for determining the state of a field, comprising:

2. The condition determination method according to claim 1 , further comprising the step of calculating an amount of microorganisms to be spread on the soil from the estimated microbial activity.

3. The condition determination method according to claim 2 , further comprising the step of outputting a map showing the application amounts at a plurality of points in the field.

4. further comprising a step of acquiring soil environment information of the soil; the estimating step includes estimating the microbial activity in the soil from the acquired gas abundance and soil environmental information based on a previously prepared relationship between the gas abundance, soil environmental information, and microbial activity; The state determination method according to claim 1 .

5. The condition determination method according to claim 4 , wherein the soil environment information of the soil includes any one selected from the group consisting of temperature, moisture content, oxygen concentration in the soil surface or in the soil, and pH value.

6. The gas is N ２ O, N ２ , NO, CH ４ , and CO ２ The state determination method according to claim 1 , wherein the state determination method includes any one selected from the group consisting of:

7. The condition determination method according to claim 1 , wherein the step of estimating includes estimating the microbial activity in the soil from the acquired gas using machine learning or multivariate analysis.

8. the acquiring step includes acquiring, at a plurality of locations in the field, gas abundances of the gas associated with positions in the field; the estimating step includes estimating the microbial activity at the plurality of locations from the plurality of acquired amounts of gas present based on a relationship between the amount of gas present and the microbial activity prepared in advance; the calculating step includes calculating the amount of microorganisms dispersed at the plurality of locations from the plurality of estimated degrees of microbial activity; The condition determination method further includes a step of mapping the calculated application amount in association with a position in the field. The state determination method according to claim 2 .

9. a calculation circuit and a storage device that stores a relationship between the amount of gas present and the microbial activity; The arithmetic circuit comprises: Obtaining the amount of gas present in the surface soil or in the soil measured by a sensor placed in the soil in the field; estimating a microbial activity in the soil from the acquired gas abundance based on the relationship stored in the storage device; The state determination device is configured to perform the following.

10. a sensor for measuring the amount of gas present in the surface soil or in the soil in a field; A placement device including a placement mechanism capable of placing the sensor in the soil and a moving device capable of moving in the field; The state determination device according to claim 9 ; a sensor management device including a cleaning device capable of removing dirt adhering to the sensor and a calibration device capable of calibrating the sensor; A state determination system comprising:

11. A method for learning an estimation model executed by an arithmetic circuit, comprising: obtaining gas abundances of gases measured in and over the soil surface; Obtaining the microbial activity of the soil; a step of learning the estimation model that estimates the microbial activity from the amount of gas present by performing learning using training data including the acquired microbial activity as input data and the acquired amount of gas present as output data; including, learning methods.

Images