Agricultural chemical management system, agricultural chemical management method, and agricultural chemical management program

Abstract

To allow optimal agricultural chemicals to be selected from the viewpoint of safety in each of areas.SOLUTION: An agricultural management system for proposing application of agricultural chemicals includes: a model storage unit which stores, for each of agricultural chemicals and modes of application thereof, an evaluation model in which region-specific regional information is associated with an evaluation value according to the risk of application of the agricultural chemicals in the region; an evaluation unit which calculates, on the basis of the evaluation models and target region information which is regional information of a target region, the evaluation value in the target region, for each of agricultural chemicals and the modes of application thereof; and a proposal unit which proposes application of the agricultural chemicals, on the basis of the calculated evaluation values. The evaluations values are one or more evaluation values according to at least two of a consumer risk, an exposure risk, and a biological impact risk.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a pesticide management system, a pesticide management method, and a pesticide management program. [Background technology]

[0002] The United Nations adopted the 2030 Agenda for Sustainable Development, which includes the Sustainable Development Goals (SDGs). Following the adoption of this agenda, countries around the world are calling for a reduction in the use of pesticides. For example, Japan's Green Food System Strategy outlines measures to reduce the environmental burden, such as reducing the amount of chemical pesticides used in Japan by 50% by 2050. In addition, the European Union's Farm to Fork Strategy proposes reducing the use of chemical pesticides by 50% by 2030. In this way, there is a demand for reducing the amount of chemical pesticides used and for safe use. Patent Document 1 describes a system for searching for candidate pesticides, and Patent Document 2 describes a system for supporting a user in pest control operations. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-9587 A [Patent Document 2] JP 2023-84773 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in Patent Document 1 and Patent Document 2, safety risks were not fully considered. Patent Document 1 describes toxicity data for each pesticide, but does not evaluate the risks of applying the pesticide in each region, and therefore could not evaluate the risks in each region. In this way, the conventional system was unable to apply pesticides optimally from the viewpoint of safety in each region.

[0005] The present invention has been made in consideration of the above points, and provides a pesticide management system, a pesticide management method, and a pesticide management program that enable pesticides to be applied optimally in each region from the standpoint of safety by conducting local risk assessments of pesticides. [Means for solving the problem]

[0006] (1) The present invention has been made to solve the above-mentioned problems. One aspect of the present invention is a pesticide management system that proposes the application of a pesticide, and includes: a model memory unit that stores an evaluation model in which, for each pesticide and its application mode, regional information specific to the region is associated with an evaluation value corresponding to the risk of applying the pesticide in that region; an evaluation unit that calculates the evaluation value in a target region for each pesticide and its application mode based on target region information, which is the regional information of the target region, and the evaluation model; and a proposal unit that proposes the application of the pesticide based on the calculated evaluation value, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk.

[0007] (2) Another aspect of the present invention is a pesticide management method that suggests the application of a pesticide, which stores an evaluation model in which, for each pesticide and its application mode, region-specific regional information and an evaluation value corresponding to the risk of applying the pesticide in the region are associated with each other, and calculates the evaluation value in a target region for each pesticide and its application mode based on target region information, which is the regional information of the target region, and the evaluation model, and suggests the application of the pesticide based on the calculated evaluation values, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk.

[0008] (3) Another aspect of the present invention is a pesticide management program that causes a computer of a pesticide management system that proposes the application of pesticides to execute a reading procedure for reading out an evaluation model in which, for each pesticide and its application mode, region-specific regional information and an evaluation value corresponding to the risk of applying the pesticide in that region are associated with each other; a risk calculation procedure for calculating the evaluation value in a target region for each pesticide and its application mode based on target region information, which is the regional information of the target region, and the evaluation model; and a proposal procedure for proposing the application of the pesticide based on the calculated evaluation value, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk. Effect of the Invention

[0009] According to the present invention, it is possible to select an optimum pesticide from the viewpoint of safety for each region. [Brief description of the drawings]

[0010] [Figure 1] 1 is a schematic diagram showing an overview of an embodiment of the present invention. [Diagram 2] FIG. 4 is a schematic diagram showing an example of a screen flow according to the embodiment. [Diagram 3] FIG. 11 is a schematic diagram showing another example of a screen flow according to the embodiment. [Figure 4] 1 is a schematic diagram showing an example of a pesticide management system according to an embodiment of the present invention. [Diagram 5] FIG. 2 is a sequence diagram showing an example of a process of the pesticide management system according to the present embodiment. [Figure 6] FIG. 2 is a schematic block diagram showing the configuration of a learning server according to the present embodiment. [Figure 7] FIG. 2 is a schematic diagram showing an example of region-related information according to the embodiment; [Figure 8] FIG. 4 is a schematic diagram illustrating an example of first learning data according to the present embodiment. [Figure 9] FIG. 4 is a schematic diagram illustrating an example of second learning data according to the embodiment. [Figure 10] FIG. 4 is a schematic diagram illustrating an example of first additional learning data according to the present embodiment. [Figure 11] FIG. 11 is a schematic diagram illustrating an example of second additional learning data according to the present embodiment. [Figure 12] FIG. 4 is a flow diagram showing the processing of a learning server according to the embodiment. [Figure 13] FIG. 2 is a schematic diagram showing an example of a risk assessment according to the present embodiment. [Figure 14] FIG. 1 is a schematic diagram showing an example of a spraying device according to an embodiment of the present invention. [Figure 15] FIG. 2 is a schematic block diagram showing the configuration of a user terminal according to the present embodiment. [Figure 16] FIG. 2 is a schematic block diagram showing the configuration of a management server according to the present embodiment. [Figure 17] 4 is a schematic diagram showing an example of user basic information according to the embodiment; FIG. [Figure 18] FIG. 2 is a schematic diagram showing an example of cultivation history information according to the present embodiment. [Figure 19] FIG. 2 is a schematic diagram showing an example of disease history information according to the present embodiment. [Figure 20] FIG. 2 is a schematic diagram showing an example of region-related information according to the embodiment; [Figure 21] FIG. 4 is a schematic diagram showing an example of inventory information according to the embodiment; [Figure 22] 2 is a schematic diagram showing an example of spraying equipment information according to the present embodiment. FIG. [Diagram 23]4 is a schematic diagram showing an example of scattering reservation information according to the embodiment; FIG. [Figure 24] FIG. 2 is a schematic diagram showing an example of control history information according to the present embodiment. [Diagram 25] FIG. 1 is a schematic block diagram showing the configuration of a drone according to an embodiment of the present invention. [Figure 26] FIG. 11 is a sequence diagram showing an example of a process related to a pesticide management application according to the present embodiment. [Figure 27] FIG. 11 is a sequence diagram showing an example of another process related to the pesticide management app according to the embodiment. [Figure 28] FIG. 2 is an explanatory diagram illustrating the hardware configuration of each device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] (First embodiment) Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The pesticide management system S1 according to this embodiment selects the most appropriate pesticide for that time and place by performing local risk assessment of registered pesticides in real time in response to the occurrence or prediction of disease damage. This allows the pesticide management system S1 to reduce the amount of pesticide sprayed, minimizing the impact on people and the environmental load.

[0012] An assessment model is used for risk assessment. The assessment model is a trained model generated by machine learning based on information such as physicochemical properties, animal / plant metabolism, environmental fate, toxicity, ecotoxicity, and crop / soil residues, which are written in the test reports for pesticide registration in each country / region and the evaluation reports of tests by the government, which are read into AI (Artificial Intelligence). The risk assessment is carried out based on the information written in the test reports and evaluation reports, taking into account the climate, soil quality, etc. specific to the region, as well as the most recent weather, and the optimal pesticide, spray date, spray amount, spray area, spray method, etc. are determined.

[0013] Based on the evaluation results, the pesticide management system S1 works in cooperation with the drone base to automatically (including some manual) spray pesticides using agricultural machinery such as drones. In this way, the pesticide management system S1 optimizes the timely spraying of pesticides, saves labor, and reduces the environmental impact. In this way, the pesticide management system S1 reduces the amount of pesticide applied, reduces the risk of human exposure, and reduces the risk of ecological impacts.

[0014] [Outline of this embodiment] FIG. 1 is a schematic diagram showing an overview according to an embodiment of the present invention. For example, when a disease occurs or is predicted to occur in farmland managed by a specific user, user disease information specific to that user is extracted. The user disease information includes the disease, the area in which the disease occurs, the target crop affected by the disease, and the scale of damage. The user disease information also includes neighboring crops of the target crop, or the control history at the location of the disease, but this information does not have to be included. The location of the disease is a representative point of the area in which the disease occurs.

[0015] When the scale of damage is included, the risk assessment value will change depending on the scale of damage. In this case, different pesticides and application methods will be used in areas of farmland where damage is greatest and in their surrounding areas. Pesticides with higher safety / lower ecological impacts will be selected for areas surrounding farmland. By using highly effective pesticides (with safety concerns) only inside farmland, the distance to water sources will be longer, reducing the risk of ecological impacts outside the farmland. When neighboring crops are included, a pesticide that causes little or no damage to the neighboring crops is selected.

[0016] The management server 3 of the pesticide management system S1 displays the input user disease information on the user terminal 2. The user terminal 2 receives input of the pesticide spraying area for the disease in the user disease information from the user. The management server 3 acquires the soil quality, topography, and climate of the spraying location based on the spraying location, which is a representative point of the spraying area, and the area-related information. The area-related information is information in which the location, soil quality, topography, and climate are previously associated. The management server 3 also acquires weather information representing the weather at the spraying location from an external server 4 that stores the location and the weather in association with each other. The weather information is, for example, information indicating the amount of rainfall, and indicates the amount of rainfall per week, day, or hour for a specific period. The weather information may be information indicating the weather, the position of rain clouds, temperature, humidity, wind, warnings and advisories, or the amount of ultraviolet light. The management server 3 generates target area information that indicates the soil type, topography, climate, and weather of the spraying location. In this way, the management server 3 generates target area information that is local information from the spraying location. The management server 3 also generates weather information at each time.

[0017] The management server 3 stores an evaluation model that associates area-related information with an evaluation value of the risk evaluation of the pesticide for each pesticide and each application mode of the pesticide. The application mode of the pesticide is, for example, the amount of the pesticide sprayed (one example of the amount of the pesticide applied). The application mode of the pesticide may include a spraying area or a spraying method. The spraying area is represented by position information by a GPS (Global Positioning System) and indicates the spraying location and range where the pesticide is sprayed. The spraying area may be input by a user or may be automatically determined from the area where the disease occurs. The spraying method may include an average spraying amount per unit area, equipment or parts thereof used for spraying the pesticide (for example, the type of nozzle or sprayer), a spraying path, a spraying speed, a spraying height, a spraying direction, or a spraying method. The spraying method may include the size and type of spray droplets, a spray pattern (cross-sectional shape of the spray), a spray angle, a flow rate, or a spraying timing.

[0018] The assessment value for risk assessment is a value according to the risk posed by the application of the pesticide in the relevant area, with a higher value being assigned to lower risks. There are three types of risk: consumer risk, exposure risk, and ecological impact risk, and the assessment value is calculated as one or more values ​​based on these risks. Consumer risk assessment (ADI, ARfD) evaluates the health risk when consumers are exposed to the pesticide from crops. Worker exposure assessment evaluates the risk of exposure for workers who handle the pesticide. Worker exposure assessment may also include an assessment of the risk of exposure for non-workers, such as passersby. Ecological risk assessment evaluates the impact of the pesticide on the environment. Assessment values ​​are based on physicochemical properties, environmental fate, predicted concentrations, predicted environmental concentrations, human effects, or ecological effects.

[0019] The assessment model is generated by machine learning based on pesticide test information, which indicates the information described in the test reports for pesticide registration in each country / region and the evaluation reports of tests by the government. The pesticide test information includes the test site and the risk assessment value of the pesticide for each pesticide and its application mode (application amount, etc.). Here, the learning server 1 of the pesticide management system S1 extracts test area information that indicates the soil quality, topography, and climate of the test site using the area-related information. The learning server 1 uses the extracted soil quality, topography, and climate of the test site and the evaluation value as learning data, and generates an evaluation model by performing machine learning using the learning data. This enables the assessment model to calculate local assessment values, such as risk assessment values ​​for pesticides in each region, from regionally-specific information such as the soil type, topography, and climate of each region, based on global and diverse pesticide testing information from the government. The half-life differs depending on the soil type, so the evaluation value may also change. The distance to the water source and the slope of the terrain may change, so the evaluation value may also change.

[0020] The management server 3 stores pesticide-related information in which the target crop, pest, and scale of damage that is affected by the disease are previously associated with the pesticide and the application mode of the pesticide for controlling the disease. When the target crop, pest, and scale of damage indicated by the user disease information are input, the management server 3 selects the pesticide and the application mode of the pesticide for controlling the disease of the user disease information based on the pesticide-related information. The selected pesticide and the application mode of the pesticide are also referred to as the candidate pesticide and the candidate application mode.

[0021] The management server 3 inputs target area information into the assessment model for each candidate spraying date, which is a candidate date for spraying, the candidate pesticide, and the candidate application mode, and outputs an assessment value for risk assessment of the pesticide. As described above, the management server 3 stores an assessment model that can calculate the risk assessment value of pesticides in each region using the region-related information, and inputs the target region information converted from the spraying location to this assessment model using the region-related information. This allows the management server 3 to calculate the risk assessment value adapted to the location where the disease occurred (the pesticide spraying location). The candidate spraying dates may be input by the user, may be automatically selected, or may be selected from a desired spraying period input by the user. If the target area information includes weather information, the weather information indicates the candidate spraying date and the amount of rainfall for a specified period thereafter (e.g., one week). For example, if the sprayed pesticide is washed away by rain before it dries, the pest control effect may be reduced and the washed-away pesticide may increase various risks in the risk assessment. The management server 3 calculates the evaluation value according to the weather information based on the candidate spraying dates, so that it can propose the optimal timing as the spraying date that reduces the risks of the pesticide while preventing a reduction in the pest control effect.

[0022] The management server 3 selects the candidate application date, candidate pesticide, and candidate application mode as an application plan based on the evaluation value, and outputs a list of application plans as proposal information. The application plan with the highest evaluation value is selected. The application plan also includes the application area. This allows the management server 3 to propose a pesticide application with a low risk assessment, that is, the application date, pesticide, application amount, application area, application method, etc. In other words, the pesticide management system S1 can propose an optimal pesticide application to reduce consumer risk, exposure risk, or ecological impact risk.

[0023] When a spraying request based on an application plan selected by a user (also referred to as an "implemented application plan") is input, the drone is loaded with a container filled with a pesticide formulation (including a mixture) of the pesticide indicated in the implemented application plan. The drone is deployed on the application date indicated in the implemented application plan and sprays the pesticide formulation at the spraying location indicated in the spraying area using the spraying method indicated in the application plan. As a result, the pesticide management system S1 can optimize pesticide application by timely spraying, reduce labor, and reduce environmental impact, as well as reduce the amount of pesticide applied, reduce the risk of human exposure, and reduce the risk of ecological impact.

[0024] [Screen Flow] FIG. 2 is a schematic diagram showing an example of a screen flow according to the present embodiment. Screen G10 is a top page displayed on each user terminal 2, and shows a state in which a user with the user name "Mr. A" is logged in. The user is, for example, a farm manager, and is a user who uses the service. The statement G101 is a disease notification that warns that a disease has occurred in the farmland managed by the user A. For example, the monitoring device 53-n monitors the farmland of the user A and predicts that a disease has occurred based on a captured image, and this disease notification is notified and displayed. The statement G101 may be a reminder of the planned date of control, or may be a notification by an icon or the like. When the statement G101 is tapped, the screen G21 is displayed.

[0025] The button BT101 is a button for transitioning to a disease confirmation page for confirming diseases on the farmland of the user A. When the button BT101 is tapped, a screen G21 is displayed. Button BT102 is a button for transitioning to a pesticide selection page where the user selects candidate pesticides for the user A's farmland. When button BT102 is tapped, screen G11 is displayed. In addition, screen G10 has a button for transitioning to a schedule page for checking the farmland and crop management schedule and the pesticide spraying schedule for user A's farmland. Screen G10 has buttons for transitioning to pages showing the growth status of the crops and the history of pesticide spraying, risk assessment for each crop and each pesticide spraying, and applying for pesticide spraying. There are buttons for transitioning to pages of agricultural examples and agricultural SNS (social network service). The agricultural examples are, for example, examples based on the growth status of the crops and the history of pesticide spraying.

[0026] Screen G11 is a pesticide selection page where the user selects a candidate pesticide. In screen G11, the user inputs items for selecting the spraying purpose and the pesticide candidate. For example, the user selects disease prevention, control, weeding, or growth regulation as the spraying purpose. The user also selects a disease from a list to select a pesticide candidate. The user can also select a pesticide directly. After the user has selected the spraying purpose and the pesticide candidate, if button BT12 is tapped, screen G22 is displayed.

[0027] Screen G21 is a disease confirmation page where the user confirms diseases. When transitioning to screen G21 by tapping sentence G101, the user confirms the disease that has occurred on screen G21. In this specification, "occurred" includes a high possibility of occurrence. When transitioning to screen G21 by tapping button BT101, the user confirms the disease that has occurred or is predicted to have occurred on screen G21. In this case, the date of occurrence of the disease may be displayed on screen G21. In this specification, "occurred or occurred" is referred to as "occurred". The screen G21 displays, for example, an occurrence area E11 showing the location where the disease occurred, a target crop showing the crop where the disease occurred, and the disease that occurred. When the user wants to control the disease confirmed on the screen G21, the user taps the button BT21. When the button BT21 is tapped, the screen G22 is displayed.

[0028] Screen G22 is a pesticide spraying page for setting the spraying of pesticides. On screen G22, the user sets the spraying area for spraying the pesticide and the spraying schedule for spraying the pesticide. For example, when spraying the pesticide over the entire farmland A including the outbreak area, the user sets the spraying area E22. The spraying area may be selected by the user swiping or tapping. For example, the user may select an area on the screen as the spraying area by swiping to surround the area or by tapping the apex of the area. In addition, if the user selects an area other than farmland managed by the user (for example, a road or farmland managed by someone else) as the spraying area, a warning may be displayed on screen G22, or the spraying area may be limited to farmland managed by the user. The user sets, for example, a period during which the user wishes to spray the pesticide as the spraying schedule. The user may be able to specify the spraying date or the spraying date. In addition, the spraying schedule may be input as a default date prior to the occurrence date of the disease, or the occurrence date may be displayed near the spraying schedule. After the user inputs the spraying area and the spraying schedule, when button BT22 is tapped, screen G23 is displayed.

[0029] The screen G23 is a plan selection page for selecting an application plan (an example of proposed information) for spraying a pesticide. On the screen G23, the user selects an application plan from a list G231 of application plans. Each application plan includes information on an evaluation value (score) of a risk assessment due to spraying of the pesticide, a pesticide formulation representing the pesticide to be sprayed, a spraying date for spraying the pesticide, a spraying amount of the pesticide, a spraying area, or a spraying method. The user can narrow down the application plans in the list G231 by tapping the button BT231 and specifying a narrowing-down condition on a screen that is displayed in a pop-up. For example, the user specifies the condition of the evaluation value, the pesticide formulation, the spraying date, the spraying amount, the spraying area, or the spraying method as the narrowing-down condition. After the user selects an application plan, if the button BT232 is tapped, the screen G24 is displayed.

[0030] The screen G24 is a spray confirmation page for confirming the contents of the pesticide spray. The user confirms the contents of the selected application plan in the display G241 of the screen G24. When the button BT241 is tapped, a spray request based on the implementation application plan selected by the user is generated. On the screen G24, for example, another application plan that increases the evaluation value of the risk assessment may be proposed. When changing to the proposed other application plan, the user taps the button BT242. In this case, the user checks the contents of the changed other application plan on the display G241 of the screen G24, and when the button BT241 is tapped, a spraying request based on the other application plan is generated. Also, in the screen G24, a check box may be displayed for the user to select whether or not to notify when another application plan can be proposed. For example, the weather information changes after the user decides on an application plan, and the evaluation value of each application plan increases or decreases depending on the weather information. When an application plan with a higher evaluation value than the application plan selected by the user occurs due to a change in the weather information, the user who selected to notify receives the notification. When the notification is tapped, the screen G23 is displayed, and the implementation plans are displayed in the list G231 in order of the highest evaluation value. The user can change to another application plan by selecting another application plan with a higher evaluation value than the application plan selected in the past and then tapping the button BT232.

[0031] FIG. 3 is a schematic diagram showing another example of a screen flow according to the present embodiment. Screen G22-1 and screen G22-2 are examples of screen G22 in FIG. The map G221-1 and the map G221-2 display farmland owned by the user as map information. Each location on the farmland is colored according to the probability of disease occurrence, and a heat map is displayed on the map. The map G221-1 and the map G221-2 can be enlarged or reduced, and the display position can be changed. The map G221-1 and the map G221-2 may be displayed in three dimensions, with altitude or the probability of occurrence as one axis.

[0032] The spraying area E22 in Fig. 2 can be narrowed or widened. When the user narrows the range of the spraying area E22-1 on screen G22-1 in Fig. 3, the area is changed to a spraying area E22-2 narrower than the spraying area E22-1, for example, as in screen G22-2. In this case, the spraying area of ​​the pesticide becomes narrower, so the user can reduce the risk of the pesticide, and the evaluation value of the risk assessment becomes higher. Conversely, when the user expands the range of the spraying area E22-2 on the screen G22-2, the spraying area is changed to a spraying area E22-1 that is wider than the spraying area E22-2, as in the screen G22-1, for example. In this case, it may be possible to prevent the occurrence of disease in crops in areas where disease has not occurred.

[0033] In addition, in the screen G22-1, the speech bubble suggests narrowing the scattering area E22-1. When the user taps the speech bubble, the scattering area E22-2 may be automatically selected. For example, the scattering area E22-2 may be automatically selected based on the occurrence probability, for example, as an area where the occurrence probability is equal to or greater than a threshold. The scattering area E22-2 may be automatically selected based on the occurrence probability and the evaluation value, for example, as an area including an area where the occurrence probability is equal to or greater than a threshold and where the evaluation value is the smallest.

[0034] [System Configuration] FIG. 4 is a schematic diagram showing an example of a pesticide management system S1 according to the present embodiment. The pesticide management system S1 includes an administrative server P, a learning server 1, a user terminal 2-n (n is a natural number; also referred to as user terminal 2), a management server 3, an external server 4, a base 51, a drone 52, and a monitoring device 53-m (m is a natural number; also referred to as monitoring device 53). The base 51, drone 52, and monitoring device 53 are also referred to as spraying equipment 5. The spraying equipment 5 is provided, for example, for each region (for example, an organization such as an association or a corporation), and is shared by multiple users. The drone 52 may be a vehicle or an airplane capable of spraying a pesticide formulation.

[0035] The administrative server P is a server operated by a public institution of each country / region, and publishes test reports for pesticide registration in each country / region, and evaluation reports of tests by the administrative side. The administrative server P accepts applications for spraying pesticide formulations from each user. If the application is approved, the user can spray the pesticide formulation in the application mode requested. The learning server 1 acquires pesticide test information indicating the information contained in the test reports and evaluation documents published by the administrative server P. The learning server 1 generates an evaluation model by performing machine learning based on the pesticide test information and region-related information. Furthermore, when the same evaluation value as that in the test report or evaluation report is measured after the pesticide formulation is sprayed, the learning server 1 may acquire additional learning data including the sprayed pesticide formulation, the soil quality, topography, and climate of the spraying location (e.g., the center position of the spraying area), and the measured measurement value or predicted value. The learning server 1 performs additional machine learning on the trained first evaluation model based on the additional learning data. The learning server 1 may perform new machine learning based on the additional learning data to generate another first evaluation model. The first evaluation model may be combined with the second evaluation model to form an evaluation model.

[0036] The user terminal 2 is a user terminal of each user who manages farmland to which pesticides are sprayed. For example, the user terminal 2-1 is a user terminal of a user who manages farmland E1. The user terminal 2-2 is a user terminal of a user who manages farmland E2. The management server 3 acquires and stores the evaluation model from the learning server 1. The management server 3 acquires monitoring result information from the monitoring device 53. The monitoring result information includes user disease information including disease, disease occurrence area, target crop affected by disease, and damage scale, and the like, and the probability of disease occurrence. When the spraying area and spraying schedule are input from the user terminal 2, the management server 3 acquires external information such as weather information from the external server 4 based on the spraying location that is a representative point of the spraying area. The management server 3 inputs the user disease information and target area information including weather information into the evaluation model and calculates an evaluation value for each application plan. The management server 3 outputs the application plan based on the evaluation value to the user terminal 2 as proposal information.

[0037] When the user selects an application plan, the management server 3 acquires a spraying request from the user terminal 2. The management server 3 generates spraying application information to be used in an application for spraying the pesticide formulation based on the selected application plan (implementation application plan). The user terminal 2 generates a spraying application form from the generated spraying application information, and transmits the spraying application form to the administrative server P to apply for spraying the pesticide formulation. The spraying application information and the spraying application form include the implementation location, the planned implementation date, the crop name, the name of the pesticide to be sprayed, the application amount or dilution factor per 10a (ares: legal measurement unit of area), and the like. Here, the implementation location is location information based on the spraying area, the planned implementation date is the spraying date, the crop name is the target crop, the name of the pesticide to be sprayed is the pesticide formulation, and the application amount or dilution factor per 10a (ares: legal measurement unit of area) is generated based on the spray amount. The application for spraying the pesticide formulation may be made by the user mailing or bringing a paper on which the spraying application form is printed.

[0038] The management server 3 acquires inventory information of the pesticide formulation stored by the base 51 and determines whether or not there is a pesticide formulation equal to or greater than the amount indicated in the implementation application plan. If there is no pesticide formulation equal to or greater than the amount to be sprayed, the management server 3 orders the pesticide formulation. When spraying of the pesticide formulation is permitted and there is more pesticide formulation than the amount to be sprayed, the management server 3 issues a spraying instruction to the drone 52 when the spraying date indicated in the implementation application plan arrives. The drone 52 is a spraying device that sprays the pesticide formulation based on the implementation application plan. The drone 52 transmits spraying performance information to the management server 3.

[0039] [Sequence configuration diagram] FIG. 5 is a sequence diagram showing an example of the processing of the pesticide management system S1 according to this embodiment. (Step S101) The learning server 1 acquires pesticide test information that the administrative server P makes public. (Step S102) The management server 3 acquires the trained model generated based on the pesticide test information in step S101.

[0040] (Step S110) The spraying equipment 5 (monitoring device 53-n) transmits the monitoring result information to the management server 3. (Step S111) If the disease occurrence probability included in the monitoring result information in step S110 is equal to or greater than a threshold, the management server 3 transmits a disease notification to the user terminal 2 of the user who manages the farmland in the occurrence area. (Step S112) The user terminal 2 transmits to the management server 3 the spray setting information including the spray area and the spray schedule based on the input from the user. (Step S113) The management server 3 acquires external information such as weather information from the external server 4 based on the spraying location that is the representative point of the spraying area in step S112.

[0041] (Step S114) The management server 3 generates one or more application plans based on the user disease information included in the monitoring result information in step S110 and the target area information including the acquired weather information, and transmits the application plans to the user terminal 2. (Step S115) The user terminal 2 transmits to the management server 3 a spraying request including the implementation application plan selected by the user. (Step S116) The management server 3 performs the following based on the implementation plan of step S115: A reservation is made to spray pesticides using the spraying equipment 5 (base 51 and drone 52).

[0042] (Step S117) The spraying equipment 5 (base 51) transmits inventory information of the stored pesticide formulations to the management server 3. The inventory information includes the amount of the pesticide formulation (excluding the pesticide formulation to be sprayed) for each pesticide formulation. The management server 3 determines whether or not the amount of the pesticide formulation in the inventory information in step S117 for the pesticide formulation in the implementation application plan of step S115 is equal to or greater than the amount to be sprayed on the spraying date. (Step S118) If the amount of the pesticide formulation is not enough to cover the application amount, the management server 3 transmits order information for the pesticide formulation to the external server 4. This external server 4 is a server of a pesticide manufacturer or a pesticide retailer. In this order, the desired delivery date is set to a date prior to the application date. (Step S119) The external server 4 notifies the spraying equipment 5 (base 51) of the scheduled delivery date. The pesticide formulation ordered in step S118 is delivered to the base 51 on the scheduled delivery date. If the scheduled delivery date is later than the scheduled spraying date, the management server 3 may notify the user terminal 2 and propose an application plan for a spraying date after the scheduled spraying date or an application plan with a modified pesticide formulation, based on the evaluation value.

[0043] (Step S120) The management server 3 generates application information to be used for applying for the spraying of the pesticide formulation based on the implementation application plan of step S115, and transmits the information to the user terminal 2. (Step S121) The user terminal 2 transmits to the administrative server P a spraying application form generated from the spraying application information in step S120. (Step S122) The user terminal 2 receives from the administrative server P the application result for the spraying application form transmitted in step S121. (Step S123) The user terminal 2 transmits the application result received in step S121 to the management server 3. If the application result in step S123 is permission for spraying of pesticides, the management server 3 may perform the operations of steps S116 to S119, or steps S118 and S119.

[0044] (Step S131) ​​If the application result in step S123 is to permit spraying of the pesticide, the management server 3 transmits a spraying instruction to the spraying equipment 5 (base 51 and drone 52) on the spraying date indicated in the implementation application plan in step S115. (Step S132) The drone 52, at the base 51, mounts a container filled with the pesticide formulation indicated in the implementation application plan of step S115, and sprays the pesticide in the application mode indicated in the implementation application plan of step S115. When spraying the pesticide, the drone 52 repeatedly transmits spraying information to the management server 3. (Step S133) The management server 3 reports the spray status to the user terminal 2 based on the spray information of step S132.

[0045] (Step S134) When spraying is completed, the drone 52 generates spraying record information and transmits it to the management server 3. (Step S135) The management server 3 generates a spraying history report based on the spraying history information and transmits it to the user terminal 2.

[0046] (Step S141) When an evaluation value is measured for the spraying of the spraying record information in step S135, the user terminal 2 transmits the input evaluation value to the management server 3. (Step S142) The management server 3 identifies the spraying location from the spraying area indicated by the spraying record information, and generates a measurement result report based on the region-related information, including the pesticide formulation sprayed, the application mode of the pesticide formulation (including the name of the region where it was sprayed), weather information on the day of spraying obtained from the external server 4, and an evaluation value. The management server 3 transmits the measurement result report to the administrative server P. The management server 3 transmits the measurement result report to the learning server 1 as additional learning data.

[0047] [Learning Server] FIG. 6 is a schematic block diagram showing the configuration of the learning server 1 according to this embodiment. The learning server 1 includes an interface I1, a memory M1, and a processor P1. The interface I1 includes a communication unit 111. The memory M1 includes a test information storage unit 121, a region-related information storage unit 122, a learning data storage unit 123, an additional learning data storage unit 124, and a trained model storage unit 125. The processor P1 includes a test information extraction unit 131, a region information acquisition unit 132, a learning data generation unit 133, a machine learning unit 134, and a setting unit 135. The learning server 1 also includes other general well-known functions of a server.

[0048] The communication unit 111 communicates with other devices. The test information storage unit 121 stores pesticide test information. The area-related information storage unit 122 stores area-related information. The learning data storage unit 123 stores the learning data. The additional learning data storage unit 124 stores the additional learning data. The trained model storage unit 125 stores the evaluation model.

[0049] The test information extraction unit 131 extracts pesticide test information from the test information storage unit 121 . The regional information acquisition unit 132 uses regional related information from the regional related information storage unit 122 to extract the soil type, topography, and climate of the test location for the pesticide test information extracted by the test information extraction unit 131 .

[0050] The learning data generating unit 133 acquires the test date and weather information after the test date of the test location included in the pesticide test information from the external server 4. The learning data generating unit 133 generates learning data including the pesticide included in the pesticide test information, the soil type, topography, and climate of the test location extracted by the regional information acquiring unit 132, the weather information on the test date and after the test date acquired from the external server 4, and the evaluation value. The learning data generating unit 133 generates first learning data including measured values ​​or predicted values ​​of various test results related to the pesticide, and second learning data in which the measured values ​​or predicted values ​​are associated with the evaluation value. The present invention is not limited to this, and the learning data may be learning data in which the evaluation value is directly associated with the pesticide, the target crop, the application mode, the test location, etc. The learning data generating unit 133 stores the generated learning data in the learning data storage unit 123.

[0051] When an evaluation value is measured for the spraying of the spraying history information, the regional information acquisition unit 132 extracts the soil quality, topography, and climate of the spraying location of the spraying history information from the regional related information storage unit 122 using the regional related information. The learning data generation unit 133 acquires the weather information of the spraying location on the spraying date and after the spraying date included in the spraying history information from the external server 4. The learning data generation unit 133 generates additional learning data including the pesticide included in the spraying history information, the soil quality, topography, and climate of the test location extracted by the regional information acquisition unit 132, the weather information of the test date and after the test date acquired from the external server 4, and the measured evaluation value. The learning data generation unit 133 stores the generated additional learning data in the additional learning data storage unit 124.

[0052] The machine learning unit 134 generates an evaluation model by performing machine learning using the learning data generated by the learning data generation unit 133. Here, the machine learning unit 134 generates a first evaluation model based on the first learning data and a second evaluation model based on the second learning data. The machine learning unit 134 generates an evaluation model in which the output value of the first evaluation model (measured value or predicted value of various test results related to the pesticide) is used as the input value of the second evaluation model. The evaluation model is a trained model that outputs an evaluation value (output value of the second evaluation model) from input information (input value of the first evaluation model) such as the pesticide, target crop, application mode, and test site. Note that the second evaluation model may be a rule-based model in which the measured value or predicted value and the evaluation value are previously associated with each other. The machine learning unit 134 stores the generated evaluation model in the learned model storage unit 125.

[0053] The machine learning unit 134 generates a pesticide prediction model and an additional evaluation model by performing machine learning using the additional learning data generated by the learning data generation unit 133. Here, the machine learning unit 134 generates a pesticide prediction model based on the first additional learning data and a second additional evaluation model based on the second additional learning data. The pesticide prediction model is a trained model that outputs pesticides and pesticide application modes from input information such as disease, target crop, and damage scale. The pesticide prediction model may be a rule-based model in which pesticides and pesticide application modes are previously associated with disease, target crop, and damage scale. The additional evaluation model is a trained model that outputs evaluation values ​​(output values ​​of the second evaluation model) from input information such as pesticides, target crops, application modes, and test sites (input values ​​of the first additional evaluation model).

[0054] Based on input from the user, the setting unit 135 sets various functions of the learning server 1. The setting unit 135 can set data items of the learning data, and can delete, for example, weather information from the data items of the learning data and the input items of the evaluation model.

[0055] 7 is a schematic diagram showing an example of region-related information according to this embodiment. As shown in the figure, it has columns for each of the items of region ID, location 1, location 2, soil type, topography, and climate. The region-related information is data consisting of rows and columns in which information is stored for each region ID.

[0056] The area ID is identification information for identifying an area. An area may be, for example, a country, a region, a prefecture, or a city, ward, town, or village, but may also be a specific area designated in advance on a map. Location 1 is the name of the region. This region name is information used to match the test location of a test report or evaluation document, for example. Location 2 is location information obtained by GPS and indicates the range of a region. This location information is used to check against areas where disease damage occurs or areas where pesticides are sprayed, for example.

[0057] The soil type indicates the typical soil type of the region. The topography shows the typical topography of the region. Climate indicates the typical climate of the region. Note that the soil type, topography, and climate for the location 2 may be information in units more detailed than the area.

[0058] 8 is a schematic diagram showing an example of the first learning data according to the present embodiment. As shown in the figure, the first learning data has columns for each of the following items: test evaluation ID, pesticide, target crop, application mode, test site, soil quality, topography, climate, weather, physicochemical properties, crop / soil residue, animal / plant metabolism, environmental fate, toxicity, and ecotoxicity. The first learning data is data consisting of rows and columns in which information is stored for each test evaluation ID. In addition, physicochemical properties, crop / soil residues, animal / plant metabolism, environmental fate, toxicity, and ecotoxicity are also referred to as test measurement information. Pesticides, test sites, application modes, and test measurement information are examples of pesticide test information.

[0059] The test evaluation ID is identification information for identifying a test or evaluation, and is used to identify test reports for pesticide registration in each country / region and evaluation reports of tests conducted by the government. The following evaluation reports are examples of test reports or evaluation documents from each country and their search pages. (1) US EPA, Pesticide Chemical Search, [online], [searched September 25, 2023], Internet <url: https: ordspub.epa.gov ords pesticides f?p="chemicalsearch:1"> (2)European Commission, Overall conclusions for Isoxaflutole, [online], [Retrieved September 25, 2023], Internet <URL:https: / / www.efsa.europa.eu / sites / default / files / consultation / consultation / Isoxaflutole_RAR.zip> (3) Food Safety Commission, Pesticide Evaluation Report Epoxyconazole, [online], [searched on September 25, 2023], Internet <URL:https: / / www.mhlw.go.jp / file / 05-Shingikai-11121000-Iyakushokuhinkyoku-Soumuka / 0000060498.pdf>

[0060] The pesticide refers to the pesticide applied in the test or evaluation. The pesticide is, for example, a pesticide formulation (including a mixture) intended for the control and prevention of disease, the removal of weeds, and the growth regulation of the target crop. The pesticide formulation may be not only liquid, but also solid or gas. Furthermore, the control and prevention of disease means, for example, the eradication and prevention of agricultural pests and diseases. The pesticide formulation intended for the growth regulation may contain fertilizer or compost. The target crop refers to the crop to which the pesticide has been applied in the test or evaluation.

[0061] The application mode refers to the mode of application of the pesticide. The application mode includes, for example, the amount of the pesticide applied in a test or evaluation, but may also include the application area or application method. The test location indicates the location where the test or evaluation was performed. The test location is, for example, the name of a region, but may also be GPS location information. The soil type, topography, climate, and weather are the soil type, topography, climate, and weather of the test site, respectively. When the region name of the test site represents a wide area, the soil type, topography, climate, and weather are representative of the soil type, average topography, climate, and weather of the region.

[0062] The physical and chemical properties are the physicochemical properties of the pesticide, and indicate the properties, stability, decomposition property, etc. of the active ingredients of the pesticide. The physical and chemical properties may also include information on the uniformity and spraying performance of the pesticide formulation, the concentration of the active ingredients of the pesticide, and changes over time in the physical and chemical properties. The physical and chemical properties include, for example, the appearance (color and shape) of the pesticide, powder size, particle size, undiluted solution stability, dilution stability or hydration, water solubility or water solubility, suspension, density, stability over time, or other measurement results required for the formulation.

[0063] Crop / soil residues indicate the substances and amounts remaining in crops or soil due to pesticide spraying, etc. Animal / plant metabolism indicates the substances metabolized in the animal or plant body and the amount of each substance. Animal metabolism provides information on the pharmacokinetics (absorption, distribution, excretion, metabolism, etc.) of the test substance. Environmental fate refers to the behavior of chemicals in environmental media such as air, water, and soil after they are released into those media, including advection within the media, movement between media, decomposition, and persistence. Toxicity is information on the toxicity of the pesticide. Toxicity includes, for example, acute toxicity, short-term toxicity, long-term toxicity, genotoxicity, carcinogenicity, reproductive toxicity, neurotoxicity, and other toxicity. Toxicity may also include the toxicity of additives or impurities that may affect the toxicity of the pesticide active ingredient. Ecotoxicity is a test to evaluate the impact of chemicals on ecosystems. An ecosystem is a material system consisting of a biological community and inorganic environments such as air, water, soil, and light. Ecotoxicity includes, for example, acute ecotoxicity, short-term ecotoxicity, medium- to long-term ecotoxicity, and other ecotoxicity. Ecotoxicity tests are conducted on mammals, birds, fish, algae, crustaceans (such as daphnia), terrestrial arthropods, microorganisms, soil organisms, plants, etc.

[0064] Pesticide test information may include information indicating exposure to pesticide users, birds, livestock, etc. Ecotoxicity pesticide test information is the measurement values ​​required to be included in test reports for pesticide registration in each country / region and in government test evaluation reports, and may include other information. Ecotoxicity pesticide test information includes test results for safety tests and food health impact assessment information. The values ​​of each item of ecotoxicity pesticide test information follow the definitions, test methods, and standards of pesticide registration or government evaluation. The pesticide test information may include classification according to crops (also called crop classification), such as crop production volume, and machine learning may be performed according to the crop classification. Performing according to the crop classification means that the training data includes information indicating the crop classification, or that a trained model is generated for each crop classification. The pesticide test information may also include information on pesticide damage and efficacy.

[0065] 9 is a schematic diagram showing an example of the second learning data according to the present embodiment. As shown in the figure, the second learning data has columns for each of the items of physicochemical properties, crop / soil residue, animal / plant metabolism, environmental fate, toxicity, ecotoxicity, consumer risk assessment, exposure assessment, and ecological impact risk assessment. The second learning data is information in which test measurement information is associated with evaluation values ​​of risk assessment (consumer risk assessment, exposure assessment, and ecological impact risk assessment). The second learning data is data consisting of rows and columns in which evaluation values ​​of risk assessment are stored for each test measurement information.

[0066] Each of the test measurement information (physicochemical properties, crop / soil residues, animal / plant metabolism, environmental fate, toxicity, and ecotoxicity) is the same item as in the first training data. A consumer risk assessment (ADI, ARfD) assesses the health risks to consumers when they are exposed to a pesticide from a crop. Exposure assessment is an assessment of the risk of exposure to humans. Worker exposure assessment is an assessment of the risk of exposure to workers who handle pesticides. Exposure assessment may be worker exposure assessment or may include worker exposure assessment as a part of it. Ecological risk assessment evaluates the impact of pesticides on the environment.

[0067] 10 is a schematic diagram showing an example of the first additional learning data according to the present embodiment. As shown in the figure, the first additional learning data has columns for each of the following items: application ID, pesticide, target crop, application mode, application location, disease, damage scale, neighboring crops, and control history. The first additional learning data is data consisting of rows and columns in which information is stored for each application ID. The first additional learning data is used to generate a pesticide prediction model by machine learning that predicts a pesticide that can prevent a disease based on a target crop, a disease, a damage scale, a neighboring crop, or a control history. The pesticide prediction model may be pesticide-related information.

[0068] The application ID is identification information for identifying application, and identifies application of a pesticide by each user (applicator). Application is, for example, application based on a spraying request or a spraying instruction. The present invention is not limited to this, and the first additional learning data may be generated based on ecotoxicity pesticide test information or other performance information.

[0069] Pesticide indicates the pesticide formulation applied. The target crop refers to an agricultural crop to which a pesticide has been applied. The application mode refers to the mode of application of the pesticide. The application mode includes, for example, the amount of application of the pesticide, but may also include the application area or application method. The application location indicates the location where the pesticide was applied. The application location is, for example, location information obtained by GPS, but may also be the name of the area.

[0070] The disease indicates a disease that has occurred, a disease that is predicted to occur, or a disease that the user wants to prevent, and is a disease that has been controlled or prevented by a pesticide. The scale of damage indicates the scale of damage caused to the target crop by the disease. Neighboring crops refer to agricultural crops that are grown near the target crop. The pest control history indicates the application history of the pesticide to the target crop, and is an application history prior to the application of the application ID.

[0071] The first additional learning data may further include items regarding efficacy and whether or not a disease can be controlled, and may be data regarding applications that were effective or applications that were able to control a disease.

[0072] 11 is a schematic diagram showing an example of second additional learning data according to this embodiment. As shown in the figure, the second additional learning data has columns for each of the following items: application ID, application date and time, applicator, pesticide, target crop, application mode, application location, soil quality, topography, climate, weather, physicochemical properties, crop / soil residue, animal / plant metabolism, environmental fate, toxicity, and ecotoxicity. The second additional learning data is data consisting of rows and columns in which information is stored for each application ID.

[0073] The application ID is identification information for identifying the application, and is the same as the identification information of the first additional learning data. The application date and time is the date and time when the pesticide was applied. The application date and time is, for example, the period during which the pesticide was sprayed (start date and time and end date and time) or the date and time when the pesticide spraying started. The applicator is identification information that indicates the user who made the spraying request. The other items are the same as those in the first learning data. Note that the soil type, topography, climate, and weather are those of the application site. The weather in the second additional learning data is the weather on the day of application and for a predetermined period of time thereafter (for example, one week).

[0074] The second additional learning data is used to generate a trained model (first additional evaluation model) that performs the same input and output as the first evaluation model. Note that the machine learning unit 134 may perform machine learning using additional learning data in which a part or all of the first additional learning data and the second additional learning data are associated with each other by an application ID. In this way, the additional evaluation model based on the first additional evaluation model and the second evaluation model can calculate an evaluation value of risk assessment based on items of the first additional learning data (e.g., neighboring crops or pest control history).

[0075] FIG. 12 is a flow diagram showing the processing of the learning server 1 according to this embodiment. This process will be described with reference to FIG.

[0076] A case where machine learning is performed based on learning data will be described. (Step S201) The test information extraction unit 131 acquires pesticide test information stored in the test information storage unit 121 by input from the administrative server P or a user. (Step S202) The regional information acquisition unit 132 extracts the soil type, topography, and climate of the test location indicated by the pesticide test information acquired in step S201. The learning data generation unit 133 generates first learning data including the soil type, topography, and climate of the extracted test location, and stores the generated first learning data and second learning data in the learning data storage unit 123. Note that the first learning data and the second learning data are stored at separate times, but may be stored at the same time.

[0077] (Step S203) The machine learning unit 134 determines whether or not an instruction to perform machine learning has been received from the user. If an instruction has been received (YES), the process of step S204 is performed. Otherwise (NO), the process of step S206 is performed. (Step S204) The machine learning unit 134 generates a first evaluation model by performing machine learning using the first learning data stored in step S202. The machine learning unit 134 generates a second evaluation model by performing machine learning using the second learning data stored in step S202. Note that the machine learning for generating the first evaluation model and the machine learning for generating the second evaluation model may be performed at separate times. (Step S205) The machine learning unit 134 combines the first evaluation model and the second evaluation model to generate an evaluation model in which the output value of the first evaluation model is used as the input value of the second evaluation model. The machine learning unit 134 stores the generated evaluation model in the trained model storage unit 125.

[0078] (Step S206) The test information extraction unit 131 judges whether the program of the flow in this figure or its own device has been stopped. If it has been stopped (YES), the flow is ended. If it has not been stopped, the process of step S201 is performed.

[0079] A case where machine learning is performed based on additional learning data will be described. (Step S201) The test information extraction unit 131 acquires the spraying record information stored in the test information storage unit 121 by input from the user. (Step S202) The regional information acquisition unit 132 extracts the soil type, topography, and climate of the application location indicated by the spraying record information acquired in step S201. The learning data generation unit 133 generates second additional learning data including the soil type, topography, and climate of the extracted application location, and stores the generated second additional learning data and first additional learning data in the additional learning data storage unit 124.

[0080] (Step S203) The machine learning unit 134 determines whether or not an instruction to perform machine learning has been received from the user. If an instruction has been received (YES), the process of step S204 is performed. Otherwise (NO), the process of step S206 is performed. (Step S204) The machine learning unit 134 generates a pesticide prediction model by performing machine learning using the first additional learning data stored in step S202. The machine learning unit 134 generates an additional evaluation model by performing machine learning using the second additional learning data stored in step S202. The machine learning for generating the pesticide prediction model and the machine learning for generating the additional evaluation model may be performed at separate times. (Step S205) The machine learning unit 134 stores the pesticide prediction model and the additional evaluation model in the learned model storage unit 125.

[0081] (Step S206) The test information extraction unit 131 judges whether the program of the flow in this figure or its own device has been stopped. If it has been stopped (YES), the flow is ended. If it has not been stopped, the process of step S201 is performed.

[0082] FIG. 13 is a schematic diagram showing an example of risk assessment according to this embodiment. Risk assessment includes an assessment based on exposure assessment and hazard assessment. Exposure assessment is based on physicochemical tests, residual tests, metabolism tests, and application amounts, and estimates the concentrations in the environment, such as soil, water, and sediment. Hazard assessment is based on ecotoxicity tests (including tests on mammals) and is carried out on representative species of each organism group. The subjects of the ecotoxicity tests include the active ingredient, formulation, metabolites and decomposition products. In this assessment, the concentration that affects the organism is evaluated.

[0083] Risk assessment can be said to be an assessment of the impact of pesticides on the environment based on the results of toxicity tests and exposure assessments. Toxicity assessment, which is the evaluation of toxicity tests, evaluates the amount of toxicity at the target risk, and exposure assessment evaluates the amount of exposure at the target risk. The evaluation value may be calculated using a predicted environmental concentration (PEC). The PEC is a value obtained by performing a calculation to predict the concentration of a pesticide in the environment, taking into account the application method of the pesticide, environmental conditions, etc., in order to evaluate the level of the pesticide present in the environment, and represents the pesticide concentration.

[0084] [Spraying equipment] FIG. 14 is a schematic diagram showing an example of the spraying equipment 5 according to this embodiment. Pesticides are stored at base 51. The pesticides are poured into container T1 (also referred to as container T1, representing containers T11, T12, T13, and T14) for each pesticide formulation. Each container T1 is affixed with a barcode that identifies the pesticide formulation and its container (or the pourable volume). For example, container T11 is filled with a specified volume of pesticide A and is affixed with spray pesticide barcode T111. Container T12 is filled with a specified volume of pesticide B and is affixed with barcode T112.

[0085] The barcode can be read by the drone 52 or a camera of each user terminal 2. The barcode is associated with a container ID that identifies the container T1. The container ID is associated with, for example, a sprayed pesticide ID, a pesticide owner, a pesticide formulation, and an equipment ID that identifies the drone 52. For example, a spraying request includes a container ID that indicates the container to be used for spraying. The drone 52 or the user terminal 2 determines whether the container ID read from the barcode of the container T1 matches the container ID included in the spraying request. The drone 52 or the user terminal 2 notifies the user of the determination result, so that the user can identify the container T1 to be used for spraying, that is, the pesticide formulation. This makes it possible to prevent the pesticide management system S1 from mistaking pesticides, etc.

[0086] [User device] FIG. 15 is a schematic block diagram showing the configuration of a user terminal 2 according to this embodiment. The user terminal 2 includes an interface I2, a memory M2, and a processor P2. The interface I2 includes a communication unit 211, an input unit 212, an output unit 213, an imaging unit 214, and a sensor unit 215. The memory M2 includes an operation storage unit 221, an application storage unit 222, and a data storage unit 223. The processor P2 includes an operation unit 23, an application unit 24, and a function unit 25. The application unit 24 includes a disease unit 241, a spray evaluation unit 242, a proposal unit 243, an inventory confirmation unit 244, and a spray confirmation unit 245. The user terminal 2 also includes general well-known functions of a mobile terminal such as a smartphone or a tablet terminal.

[0087] The communication unit 211 communicates with other devices. The input unit 212 receives input from a user. The input unit 212 is a touch panel, and receives operational input such as tapping, dragging, and flicking from the user. The output unit 213 is a display that displays information and a speaker that outputs sound. The imaging unit 214 is a camera that captures images. The sensor unit 215 is a sensor that performs sensing, such as an acceleration sensor or a fingerprint sensor that detects fingerprints.

[0088] The operation storage unit 221 stores the program of the operating system. The application storage unit 222 stores various application programs. The data storage unit 223 stores data used by the operating system and applications. The operation unit 23 executes the operating system program stored in the operation storage unit 221 using the data stored in the data storage unit 223 . The application unit 24 executes the application program stored in the application storage unit 222 by using the data stored in the data storage unit 223 . The function unit 25 also controls various other functions of the user terminal 2.

[0089] The application unit 24 executes the pesticide management application of the pesticide management system S1. When executing the pesticide management application, the application unit 24 includes a disease unit 241, a spray evaluation unit 242, a proposal unit 243, an inventory confirmation unit 244, and a spray confirmation unit 245. The disease unit 241 displays user disease information including an occurred disease or a disease predicted to occur, an area where the disease has occurred, a target crop, and a scale of damage, for the user, on the output unit 213. The disease unit 241 may predict an occurring disease based on an image of the crop captured by the user using the imaging unit 214 and a disease prediction model. The disease prediction model is, for example, a trained model that has learned the image of the crop and the occurred disease by machine learning.

[0090] The spray evaluation unit 242 displays, for the user disease information and the pesticide spraying area, a risk evaluation for each candidate pesticide for controlling the disease included in the user disease information, a candidate spraying date, and a candidate application mode. The suggestion unit 243 displays the administration plan selected based on the result of the risk assessment as proposed information. The suggestion unit 243 acquires the implementation administration plan selected by the user. The inventory confirmation unit 244 displays the inventory status of the pesticide formulation indicated in the implementation application plan based on the inventory information of the base 51. If there is no pesticide formulation for the amount required for application, the inventory confirmation unit 244 accepts an order request for the pesticide formulation indicated in the implementation application plan.

[0091] The spraying confirmation unit 245 notifies the user of the spraying date and when the drone 52 starts spraying the pesticide. When notification is selected, the spraying confirmation unit 245 displays the implementation application plan and information on the drone 52. Based on a spraying status report from the management server 3, the spraying confirmation unit 245 displays the video captured by the camera mounted on the drone 52, the position information of the drone 52, a map based on the position information, the pesticide formulation being sprayed, the spraying amount or spraying speed of the pesticide formulation, the elapsed time, and sensing information such as wind speed and temperature.

[0092] [Management Server] FIG. 16 is a schematic block diagram showing the configuration of the management server 3 according to this embodiment. The management server 3 includes an interface I3, a memory M3, and a processor P3. The interface I3 includes a communication unit 311. The memory M3 includes a user information storage unit 321, a model storage unit 322, a region-related information storage unit 323, an inventory information storage unit 324, an equipment information storage unit 325, and a control history storage unit 326. The processor P3 includes a user management unit 331, a disease management unit 332, an evaluation management unit 333, a spray adjustment unit 334, an inventory management unit 335, a spray application unit 336, and a spray management unit 337. The evaluation management unit 333 includes a target region information acquisition unit 3331, a pesticide selection unit 3332, and an evaluation unit 3333. The management server 3 also includes other general publicly known functions of a server.

[0093] The user information storage unit 321 stores user basic information including user authentication information, cultivation history information for the user to manage the crops, and disease history information for managing diseases that occur in the cultivated crops. The model storage unit 322 stores the evaluation model, the pesticide prediction model, and the additional evaluation model. The area-related information storage unit 323 stores area-related information. The inventory information storage unit 324 stores inventory information of the pesticide formulations stored at each base 51. The equipment information storage unit 325 stores spraying reservation information based on a spraying request and spraying equipment information for managing the spraying equipment 5. The control history storage unit 326 stores control history information.

[0094] The user management unit 331 performs user authentication in the pesticide management app based on the user basic information. The user management unit 331 causes the user terminal 2 of the authenticated user to display the cultivation history information or disease history information of the user. The disease management unit 332 transmits disease notification and user disease information to the user terminal 2 based on the monitoring result information from the monitoring device 53.

[0095] The evaluation management unit 333 performs risk assessment for each candidate pesticide, candidate spraying date, and candidate application mode for controlling the disease included in the user disease information, based on the user disease information and the spraying area of ​​the pesticide. Here, in the evaluation management unit 333, the target area information acquisition unit 3331 generates target area information representing the soil quality, topography, and climate of the spraying location based on the spraying location and area-related information. The pesticide selection unit 3332 selects a pesticide and an application mode based on the disease, target crop, and damage scale indicated by the user disease information and the pesticide prediction model. The evaluation unit 333 calculates an evaluation value at the spraying location based on the generated target area information and evaluation model for each application amount of the pesticide indicated by the selected pesticide and application mode. The spray adjustment unit 334 selects an application plan based on the result of the risk evaluation performed by the evaluation management unit 333. The spray adjustment unit 334 acquires the implementation application plan selected by the user from the application plan.

[0096] The inventory management unit 335 acquires inventory information of the base 51 used by the authenticated user. The inventory management unit 335 judges whether the amount of the pesticide formulation in the inventory information for the pesticide formulation indicated in the implementation application plan is equal to or greater than the application amount on the application date. If the amount of the pesticide formulation is not equal to or greater than the application amount, the inventory management unit 335 transmits ordering information for the pesticide formulation to the external server 4. The application application unit 336 generates application information to be used for applying for the application of the pesticide formulation based on the implementation application plan. The application destination is included in the user basic information. When spraying of the pesticide formulation is permitted and there is more pesticide formulation than the amount to be sprayed, the spraying management unit 337 instructs the drone 52 to spray when the spraying date indicated in the implementation application plan arrives. The spraying management unit 337 reports the spraying status to the user terminal 2 based on the spraying information received from the drone 52.

[0097] 17 is a schematic diagram showing an example of user basic information according to the present embodiment. As shown in the figure, the user basic information has columns for each item: user ID, user name, group, contract plan, managed farmland, neighboring crops, base ID, and spraying application destination. The user basic information is data consisting of rows and columns in which information is stored for each user ID, and further, neighboring crops, base ID, and spraying application destination are stored for each managed farmland.

[0098] The user ID is identification information for identifying a user. The user name indicates the user name of the user identified by the user ID. The group to which the user belongs indicates a group to which the user belongs, and is, for example, an organization such as an association or a corporation. The contract plan indicates a contract plan with which the user enters into a contract. The contract plan may be a plan according to the group to which the user belongs, or may include preferential treatment according to the group to which the user belongs. In addition, for each contract plan, the pesticide formulation that can be applied, the amount of the pesticide formulation, or the drone 52 that can be used for spraying may be determined.

[0099] The managed farmland indicates farmland managed by a user. A single user can register multiple farmlands. Neighboring crops refer to crops grown in the vicinity of managed farmland. The base ID is identification information that identifies the base 51 used when spraying pesticides on managed farmland. The distribution request destination indicates the destination to which the user sends the distribution request.

[0100] 18 is a schematic diagram showing an example of cultivation history information according to this embodiment. As shown in the figure, the cultivation history information has columns for each item of user ID, cultivation history ID, cultivated crop, cultivation start date, cultivation end date, cultivated farmland, cultivation area, cultivation location, disease history ID, pest control history ID, and crop risk assessment. The cultivation history information is data consisting of rows and columns in which information is stored for each user ID and cultivation history ID. The cultivation history information is associated with user basic information by the user ID.

[0101] The user ID is identification information for identifying a user. The cultivation history ID is identification information for identifying the cultivation of the crop by the user. The cultivated crops refer to the crops that have been cultivated by the user or that the user is cultivating. The cultivation start date indicates the date on which the user started cultivating the crop. The cultivation end date indicates the date on which the user finished cultivating the crop.

[0102] The cultivated farmland indicates the farmland on which the user cultivates the cultivated crop. The cultivation area indicates an area in which the user cultivates the crop. The cultivation area is expressed by using location information from a GPS, for example. The cultivation area may be the same as the cultivation farmland. The cultivation location indicates a representative point of the cultivation area, for example, the center of the cultivation area.

[0103] The disease history ID is identification information that is entered when a disease occurs in a cultivated crop, and is associated with the disease history information. The control history ID is identification information that is input when control or the like is performed on a cultivated crop, and is associated with the control history information. Control or the like is, for example, prevention, control, weeding, or growth regulation. The agricultural crop risk assessment is an assessment value when a risk assessment is performed on a cultivated crop. For example, when a cultivated crop is sprayed with a pesticide multiple times, the agricultural crop risk assessment is calculated based on the risk assessment values ​​calculated for each spray.

[0104] Fig. 19 is a schematic diagram showing an example of disease history information according to this embodiment. As shown in the figure, the disease history information has columns for each of the following items: user ID, disease history ID, disease start date, disease end date, outbreak area, target crop, damage scale, and control history ID. The disease history information is data consisting of rows and columns in which information is stored for each user ID and disease history ID. The disease history information is associated with user basic information by the user ID, and associated with cultivation history information by the disease history ID.

[0105] The user ID is identification information for identifying a user. The disease history ID is identification information for identifying a disease that has occurred in a crop managed by a user. The disease indicates the name of the disease or the cause of the disease (an example of disease information). The disease onset date indicates the date on which the disease occurred. The disease end date indicates the date when the disease ended.

[0106] The occurrence area indicates an area where a disease has occurred. The occurrence area is expressed using location information from a GPS, for example. The target crop indicates the target crop that is affected by the disease. The damage scale indicates the scale of damage caused by the disease. The damage scale is, for example, the diseased area where the disease occurred, the damage period, or the damage amount. The damage scale may be classified as large, medium, or small, or may be a ratio such as a disease rate obtained by dividing the diseased area by the area of ​​the cultivation area. The control history ID is identification information that is input when control or the like against disease is performed on cultivated crops, and is associated with the control history information. The disease, the area of ​​occurrence, the target crop, and the scale of damage are user disease information of the user identified by the user ID.

[0107] FIG. 20 is a schematic diagram showing an example of the area-related information according to the present embodiment. The region-related information in this figure is the same as the region-related information in Figure 7, and therefore a description thereof will be omitted. Note that in the management server 3, the position information of the location 2 obtained by GPS is mainly used for collation.

[0108] Fig. 21 is a schematic diagram showing an example of inventory information according to this embodiment. As shown in the figure, the inventory information has columns for each of the following items: base ID, container ID, pesticide formulation, capacity, remaining capacity, production date, expiration date, pesticide owner, management history, container type, weight, application schedule, spraying equipment, and spraying method. The inventory information is data consisting of rows and columns in which information is stored for each base ID and container ID. The inventory information stores data for container T1 stored at base 51.

[0109] The base ID is identification information that identifies the base 51 . The container ID is identification information for identifying the container T1 and its contents. Pesticide formulation refers to the pesticide formulation in container T1. The capacity indicates the maximum volume of liquid that can be poured into the container T1. The remaining capacity indicates the remaining capacity of the pesticide formulation in the container T1. The remaining capacity may be measured before production or transfer, or may be measured by a device in the base 51 or a drone 52, for example, by image analysis. The pesticide formulation may be transferred in a capacity less than the maximum capacity, in which case the remaining capacity at the time of transfer will be less than the maximum capacity. Also, for example, the container T1 ordered by the order request may be filled with less pesticide formulation than the maximum capacity based on the amount of spray required for spraying. This eliminates the need for excessive spraying or disposal of the pesticide.

[0110] The production date indicates the production date of the pesticide formulation in the container T1. The expiration date indicates the expiration date of the pesticide formulation in the container T1. The pesticide owner indicates the owner of the pesticide formulation in the container T1, and is, for example, the user ID of the pesticide owner. The pesticide owner may be a group to which the user belongs. If the user ID who made the spraying request does not match the user ID of the pesticide owner of the container T1, the drone 52 may not spray. In this case, the drone 52 may send a confirmation notification to the user terminal 2 of the user who made the spraying request, or may send a notification to the user terminal 2 of the pesticide owner of the container T1 to obtain permission for another user to apply the pesticide.

[0111] The management history indicates the management history of the container T1. For example, the start and end dates of production, distribution, or storage, and the locations thereof are recorded as the history. The container type indicates the type of the container T1. The container type may indicate the shape or size of the container T1, a spraying device that can be mounted on the container T1, a connector that can be connected, or a hose that can be inserted. The container type may also be the product number of the container T1. Weight indicates the weight of the container T1 and its contents. Weight may be measured before production or transfer, or may be measured by a weighing scale at the base 51 or by the drone 52. The application schedule indicates the planned date for applying the pesticide formulation in the container T1, that is, the planned spraying date. The spraying equipment indicates an equipment capable of mounting the container T1 and spraying the pesticide formulation. The spraying equipment may be an equipment type indicating the type of equipment. The spraying method indicates a spraying method that is recommended or should be followed. The spraying method may be a prohibited spraying method. For example, the user terminal 2 may read the barcode of the container T1 to calculate the container ID and display the spraying method associated with the container ID. The drone 52 may read the barcode of the container T1 to calculate the container ID and determine the spraying mode according to the spraying method associated with the container ID to spray the pesticide.

[0112] 22 is a schematic diagram showing an example of the spraying equipment information according to the present embodiment. As shown in the figure, the spraying equipment information has columns for each of the following items: equipment ID, equipment owner, affiliated base ID, equipment type, on-board container type, transportable weight, and supported spraying method. The spraying equipment information is data consisting of rows and columns in which information is stored for each equipment ID.

[0113] The device ID is identification information that identifies the spraying device (drone 52) and the monitoring device 53-n. The device owner indicates the owner of the device. The affiliated base ID indicates the base to which the device belongs. The affiliated base ID stores the base ID. The device type indicates the type of the device. For example, the device type is set to monitoring or scattering.

[0114] The equipment type indicates the type of equipment. The equipment type is, for example, the manufacturer or model number of the equipment. The equipment type may be classified according to the type and amount of the pesticide formulation that can be sprayed. The onboard container type indicates the type of container that the device can be mounted on. The transportable weight indicates the maximum weight that the equipment can transport. The supported spraying method indicates the spraying method that the equipment can support, such as the type of nozzle or sprayer for spraying pesticides.

[0115] 23 is a schematic diagram showing an example of spray reservation information according to this embodiment. As shown in the figure, the spray reservation information has columns for each of the following items: user ID, spray reservation ID, spray start date and time, spray end date and time, weather forecast, spray area, container ID, application amount, device ID, spray method, and application status. The data is made up of rows and columns in which information is stored for each user ID and spray reservation ID.

[0116] The user ID is identification information of the user who made the dispersion request. The scattering reservation ID is identification information that identifies a scattering reservation based on a scattering request. The spray start date and time is the spray date when spraying of the pesticide begins. The spraying end date and time is the planned date and time when spraying of the pesticide will end. The forecast indicates the weather on the spraying date at the time the spraying reservation is made and for a specified period thereafter. The spraying area indicates the location and range where the pesticide will be sprayed.

[0117] The container ID is identification information of the container T1 that contains the pesticide formulation used for spraying the pesticide. The planned application amount indicates the planned amount of pesticide to be sprayed. The device ID is identification information for the drone 52 used to spray pesticides. The spraying method indicates the method of spraying the pesticide. The application status indicates the application status of the pesticide, such as whether or not an application is necessary, or the application status (not yet applied, application in progress, approved), for example.

[0118] 24 is a schematic diagram showing an example of pest control history information according to this embodiment. As shown in the figure, the pest control history information has columns for each of the following items: user ID, pest control history ID, spray start date and time, spray end date and time, weather, spray area, container ID, application amount, device ID, spray method, and evaluation result. The data is made up of rows and columns in which information is stored for each user ID and spray reservation ID.

[0119] The user ID is identification information of the user who made the dispersion request. The control history ID is identification information for identifying the control performed by the user. The spray start date and time is the date when spraying of the pesticide started. The spraying end date and time is the date and time when the spraying of the pesticide was completed. The weather indicates the weather at the time the pesticide was sprayed. The spraying area indicates the location and range where the pesticide was sprayed.

[0120] The container ID is identification information of the container T1 that contains the pesticide formulation used for spraying the pesticide. The application amount indicates the amount of pesticide applied. The device ID is identification information of the drone 52 used to spray the pesticide. The spraying method indicates the method of spraying the pesticide. The evaluation result is an evaluation value of the risk assessment regarding the spraying of pesticides.

[0121] [Scattering device: Drone] FIG. 25 is a schematic block diagram showing the configuration of a drone 52 according to this embodiment. The drone 52 includes an interface I5, a memory M5, a processor P5, and a drive unit D5. The interface I5 includes a communication unit 511, an input unit 512, an output unit 513, an imaging unit 514, and a sensor unit 515. The memory M5 includes a spray reservation storage unit 521, a spray status storage unit 522, and a history storage unit 523. The processor P5 includes a flight control unit 531, an application unit 54, and a function unit 55. The application unit 54 includes an on-board setting unit 541, a flight control unit 542, a spray control unit 543, an exposure monitoring unit 544, and a spray reporting unit 545. The drive unit D5 includes a flight drive unit 561 and a spray drive unit 562. The drone 52 also includes a battery and other commonly known functions of a drone.

[0122] The communication unit 511 communicates with other devices. The input unit 512 receives input from a user. The input unit 512 is a microphone that inputs voice. The input unit 512 receives operation input from a remote controller in the case of manual operation. The output unit 513 is a speaker that outputs sound and a light that outputs light, but may also be a display or projector that outputs video. The imaging unit 514 is a camera that captures images. The sensor unit 515 is a sensor that performs sensing, such as a GPS sensor, a gyro sensor, an acceleration sensor, a magnetic sensor, an air pressure sensor, or an obstacle detection sensor.

[0123] The scattering reservation storage unit 521 stores scattering reservation information. The dispersion information storage unit 522 stores the dispersion information. The history memory unit 323 stores history information such as the flight path of the drone 52 for each time.

[0124] The flight control unit 531 controls the flight of the drone 52 by controlling the flight drive unit 561 based on the sensing information detected by the sensor unit 515. The mounting setting unit 541 mounts the container T1 based on the spray reservation information stored in the spray reservation storage unit 521. The mounting setting unit 541 also mounts the parts (type of nozzle or sprayer) for the spraying method included in the spray reservation information. When manual installation of the parts is required, the mounting setting unit 541 may notify the user terminal 2 of the parts to be installed and have the user install the parts.

[0125] The flight path unit 542 sets a spray location and a spray path on the flight path based on the spray reservation information stored in the spray reservation memory unit 521. The spray control unit 543 controls the spray drive unit 562 based on the spray reservation information stored in the spray reservation storage unit 521, thereby controlling the spraying of the pesticide.

[0126] When spraying pesticide, the exposure monitoring unit 544 uses the image captured by the imaging unit 514 to continue, stop, suspend or resume spraying of the pesticide based on whether or not there are people or animals such as workers or passersby present. When the pesticide is being sprayed, the spraying reporting unit 545 transmits spraying information to the server 3 and stores the spraying information in the spraying information storage unit 522. When the spraying of the pesticide is completed, the spraying reporting unit 545 generates spraying record information based on the spraying information stored in the spraying information storage unit 522 and transmits the spraying record information to the server 3. The functional unit 55 also controls various other functions of the drone. The flight drive 561 is a propeller and a motor. The spray drive unit 562 is a driving device for the sprayer, such as a motor.

[0127] [How the pesticide management app works] 26 is a sequence diagram showing an example of processing related to the pesticide management application according to this embodiment. The processing will be described with reference to the configurations of the user terminal 2 and the management server 3 (FIGS. 15 and 16).

[0128] (Step S301) The application unit 24 starts an application based on a user operation. (Step S302) The disease unit 241 transmits a request to refer to a disease confirmation page to the management server 3 based on a user operation. (Step S303) The disease management unit 332 transmits user disease information to the user terminal 2 based on the monitoring result information from the monitoring device 53. (Step S304) If the user requests pest control, the spraying evaluation unit 242 displays a pesticide spraying page.

[0129] (Step S305) The spray evaluation unit 242 receives input of the pesticide spraying area and spraying schedule from the user on the pesticide spraying page in step S304. (Step S306) The spray evaluation unit 242 generates spray setting information including the spray area and spray schedule input in step S305, and transmits the spray setting information to the management server 3. (Step S307) The target area information acquisition unit 3331 acquires the soil quality, topography, and climate of the spraying location based on the spraying location, which is a representative point of the spraying area in step S306, and the area-related information. The target area information acquisition unit 3331 acquires external information such as weather information from the external server 4 based on the spraying location and the candidate spraying dates included in the spraying schedule. The target area information acquisition unit 3331 generates target area information representing the soil quality, topography, and climate of the spraying location, and the weather based on the candidate spraying dates, for each candidate spraying date.

[0130] (Step S308) The pesticide selection unit 3332 inputs the disease, target crop, and damage scale indicated by the user disease information in step S303 into the pesticide prediction model stored in the model storage unit 322, and outputs one or more candidate pesticides and candidate application modes. (Step S309) The evaluation unit 3333 inputs the target area information for each candidate spraying date in step S307 for each candidate pesticide and candidate application mode in step S308 into the evaluation model stored in the model memory unit 322, and outputs the risk assessment value for each candidate spraying date, candidate pesticide, and candidate application mode.

[0131] (Step S310) The spray adjustment unit 334 selects the candidate spraying dates, candidate pesticides, and candidate application conditions as application plans in descending order of the evaluation value in step S309. The spray adjustment unit 334 generates a list of the selected application plans and transmits it to the user terminal 2. (Step S311) The proposing unit 243 displays the list of step S310, and the user selects an implementation application plan from the list. (Step S312) The proposal unit 243 generates a spraying request including the implementation application plan of step S311, and transmits the spraying request to the management server 3.

[0132] (Step S313) The spraying adjustment unit 334 generates spraying reservation information based on the implementation application plan included in the spraying request of step S312, and stores it in the equipment information storage unit 325. Here, the spraying adjustment unit 334 identifies the base ID based on the user ID and user basic information that input the spraying request. The inventory management unit 335 extracts inventory information for the identified base ID, and for the pesticide formulation indicated by the implementation application plan, determines whether the amount of the pesticide formulation in the inventory information (total amount of remaining amount) is equal to or greater than the spraying amount on the spraying date. If it is equal to or greater than the spraying amount, the inventory management unit 335 selects a container ID based on the spraying amount and the remaining amount. In addition, the spraying adjustment unit 334 identifies an equipment ID that can mount the container T1 of the container ID based on the identified base ID and spraying equipment information. The spraying adjustment unit 334 issues a spraying reservation ID, and generates and stores spraying reservation information associated with the user ID, the implementation application plan, the container ID, and the equipment ID.

[0133] If the amount of the pesticide formulation in the inventory information is less than the amount to be sprayed, the spray adjustment unit 334 generates spray reservation information without a container ID and an equipment ID. In this case, the inventory confirmation unit 244 displays a screen for the user to confirm that the pesticide formulation in the inventory information is insufficient and whether or not to order it. If the user selects "order possible," the inventory management unit 335 orders the pesticide formulation indicated in the implementation application plan. When the container ID of the container into which the ordered pesticide formulation is poured is input, the spray adjustment unit 334 identifies the container ID and the equipment ID and stores them in the generated spray reservation information.

[0134] (Step S314) The scattering application unit 336 extracts the application destination of the user's scattering application from the user basic information, and generates scattering application information based on the application destination and the scattering reservation information. The scattering application unit 336 transmits the generated scattering application information to the user terminal 2. (Step S315) The proposal unit 343 generates a spraying application form to be used for applying for spraying of the pesticide formulation based on the spraying application information in step S314, and transmits the application to the application destination.

[0135] 27 is a sequence diagram showing an example of another process related to the pesticide management app according to the present embodiment. The process will be described with reference to the configurations of the user terminal 2, the management server 3, and the drone 52 (FIGS. 15, 16, and 25).

[0136] (Step S401) At the base 51, one or more containers T1 with the container ID in the spraying reservation information are prepared. The one or more containers are filled with the pesticide formulation in the implementation specification plan in an amount equal to or greater than the amount to be sprayed. The spraying confirmation unit 245 of the user terminal 2 can read the barcode on the container T1 to identify the container ID of that container T1. The spraying management unit 337 determines whether the container ID is included in the spraying reservation information for the authenticated user ID. The spraying confirmation unit 245 displays the determination result to the user. This allows the user to determine whether the container T1 whose barcode has been read is the one to be used for spraying. Therefore, the pesticide management system S1 can prevent mistakes such as pesticide misapplication.

[0137] (Step S402) The spray management unit 337 issues a spray notification to the user terminal 2 based on the spray date (the spray start date and time of the spray reservation information) indicated in the implementation application plan. The spray notification may be a reminder before the spray date. (Step S403) When spraying of the pesticide formulation is permitted and there is more pesticide formulation than the amount to be sprayed, when the spraying date indicated in the implementation application plan arrives, the spraying management unit 337 transmits a spraying instruction to the drone 52 of the device ID of the spraying reservation information. The spraying instruction includes the spraying reservation information.

[0138] (Step S404) The loading setting unit 541 loads the container T1 based on the spray reservation information stored in the spray reservation storage unit 521. The loading setting unit 541 determines whether the container ID read from the barcode of the container T1 to be loaded or loaded at the base 51 matches the container ID included in the spray reservation information of step S403. This allows the loading setting unit 541 to determine that the container T1 to be loaded is the container T1 requested for spraying, that is, that the container T1 of the pesticide formulation in the implementation application plan has been loaded. This prevents the pesticide management system S1 from mistaking pesticides, etc.

[0139] (Step S405) The on-board setting unit 541 acquires weather information for the day and transmits it to the spraying management unit 337. The weather information is, for example, weather information in hourly units at the spraying location. The spraying management unit 337 recalculates the evaluation value of the risk assessment based on the implementation application plan, the target area information including the weather information for the day, and the evaluation model. If the evaluation value is below the threshold, the spraying management unit 337 may transmit a screen to the user terminal 2 to confirm whether or not the pesticide can be sprayed, and the spraying confirmation unit 245 may display that screen. In addition, the pipe spraying management unit 337 may re-output a list of other application plans, and the spraying confirmation unit 245 may display that list.

[0140] (Step S406) The installation setting unit 541 installs the parts (nozzle and type of sprayer) for the spraying method included in the spraying reservation information. If manual installation of parts is required, the installation setting unit 541 may notify the user terminal 2 of the parts to be installed and have the user install the parts. The spraying method may be updated based on the results of the pesticide spraying application. (Step S407) Flight path unit 542 sets a spraying location and a spraying path on the flight path based on the spraying reservation information stored in spraying reservation storage unit 521. Flight control unit 531 controls flight drive unit 561 to fly drone 52 along the flight path.

[0141] (Step S411) When spraying pesticide, the exposure monitoring unit 544 uses the image captured by the imaging unit 514 to monitor whether there are people, such as workers or passersby, or animals. (Step S412) If no people or animals are present, the spray control unit 543 controls the spraying of the pesticide by controlling the spray drive unit 562 based on the spray reservation information stored in the spray reservation storage unit 521. Specifically, the spray control unit 543 controls the average spray amount per unit area, the spray speed, the spray height, the spray direction, or the spray method based on the spray method included in the spray reservation information. For example, the spray control unit 543 controls the size and type of spray droplets, the spray pattern (cross-sectional shape of the spray), the spray angle, the flow rate, or the spray timing.

[0142] (Step S413) If the exposure monitoring unit 544 detects the presence of a person or animal in the video, it suspends the spraying of the pesticide. The exposure monitoring unit 544 notifies the user terminal 2, and may also output sound, light, or an image to the person or animal to urge them to leave the spraying area. After the suspension, if there are no more people or animals present, the exposure monitoring unit 544 resumes spraying of the pesticide. This allows the pesticide management system S1 to reduce the risk of people or animals being exposed to pesticides. The exposure monitoring unit 544 may transmit to the server 3 whether or not a person or animal was present, and if so, the time or location. The spraying management unit 337 may calculate the performance of the risk assessment value or correct the original assessment value according to the time or location. That is, the exposure monitoring unit 544 detects the presence of a person while spraying the pesticide, and the evaluation management unit 333 calculates the assessment value based on the detection result of the presence of a person. This allows the pesticide management system S1 to calculate an evaluation value according to the presence or absence of actual exposure and the duration of exposure. For example, if there is a long period of time when no people or animals are present, the exposure monitoring unit 544 increases the evaluation value of the risk assessment for the target crop (the exposure risk becomes lower).

[0143] (Step S414) The spraying reporting unit 545 transmits spraying information to the server 3 based on the control results of the flight control unit 531 and the spraying control unit 543 or the sensing information detected by the sensor unit 515, and stores the spraying information in the spraying information storage unit 522. The spraying information includes, for example, the time, the image captured by the imaging unit 514, the position information of the drone 52, the container ID of the mounted container, the pesticide formulation being sprayed, the equipped parts, the spraying method, the amount or speed of the pesticide formulation sprayed, the elapsed time, and sensing information such as wind speed and temperature, for each specific time interval. (Step S415) The spraying management unit 337 reports the spraying status to the user terminal 2. In addition to the spraying information, the spraying status report may include the disease occurrence area, the spraying area, the spraying route, a map based on the location information, and the like. The spraying confirmation unit 245 displays the information of the spraying status report. (Step S416) The spray control unit 543 determines whether or not spraying of the pesticide according to the implementation plan has been completed. If spraying has not been completed, the process of step S411 is performed again.

[0144] (Step S417) When spraying is completed, the spraying reporting unit 545 generates spraying history information and transmits it to the management server 3. The spraying history information includes the pesticide that was sprayed and the application mode of the pesticide. The spraying management unit 337 generates pest control history information based on the spraying history information and stores it in the pest control history memory unit 326. The spraying management unit 337 reports the spraying history based on the spraying history information.

[0145] As described above, in this embodiment, the pesticide management system S1 suggests the application of pesticides. The model storage unit 322 stores an evaluation model in which region-specific regional information and an evaluation value according to the risk of applying the pesticide in the region are associated with each pesticide and its application mode for each pesticide and its application mode. The evaluation management unit 333 calculates an evaluation value in the target region for each pesticide and its application mode based on target region information, which is regional information of the target region, and the evaluation model. The proposal unit 243 (or the spray adjustment unit 334) proposes the application of the pesticide based on the calculated evaluation value. The evaluation value is one or more evaluation values ​​according to at least two of consumer risk, exposure risk, and ecological impact risk. This allows the pesticide management system S1 to calculate local evaluation values, such as the evaluation value of the risk assessment of pesticides in each region, from the target area information for each region, and select the optimal pesticide for that location. This allows the pesticide management system S1 to reduce the amount of pesticide sprayed, minimizing the impact on people and the environmental load.

[0146] In this embodiment, the area-related information storage unit 323 stores area-related information associated with area information representing a location, soil quality, topography, and climate. The model storage unit 322 (an example of a pesticide information storage unit) stores a pesticide prediction model (an example of pesticide-related information) in which a target crop of a disease, disease name or disease cause information and damage scale, and a pesticide to be sprayed on the target crop and an application mode indicating the application amount of the pesticide are previously associated. The communication unit 311 (an example of an input unit) inputs a pesticide spraying location, a target crop, disease information, and damage scale. The target area information acquisition unit 3331 generates target area information representing the soil quality, topography, and climate of the spraying location based on the spraying location and the area-related information. The pesticide selection unit 3332 selects a pesticide and an application mode based on the target crop, disease information, damage scale, and the pesticide prediction model. The evaluation unit 3333 calculates an evaluation value at the application location for each of the selected pesticides and the application amounts of the pesticides indicated by the application modes, based on the generated target area information and evaluation model. In this way, the pesticide management system S1 calculates an evaluation value for a pesticide application location based on the soil quality, topography, and climate of the pesticide application location and the evaluation model, and the evaluation model corresponds to the soil quality, topography, and climate of the pesticide application location. Thus, the pesticide management system S1 can calculate an evaluation value for the risk assessment of pesticides in each area from the soil quality, topography, and climate of each area, and can select the optimal pesticide for that location.

[0147] In this embodiment, the test information extraction unit 131 extracts information on the test site where the test was conducted, consumer risk, exposure risk, and ecological impact risk for each pesticide and application amount of the pesticide in the test from the test report in the pesticide registration and the pesticide test information indicating the evaluation result of the test by the government. The regional information acquisition unit 132 (an example of a test region information acquisition unit) generates test region information indicating the soil type, topography, and climate of the test site based on the test site and the regional related information. The machine learning unit 134 generates an evaluation model by performing machine learning based on the test region information and evaluation values ​​corresponding to at least two of the consumer risk, exposure risk, and ecological impact risk. In this way, the pesticide management system S1 performs machine learning to generate an evaluation model using the test reports in the pesticide registration and the pesticide test information showing the evaluation results of the test by the government. This allows the pesticide management system S1 to generate a highly accurate evaluation model using a large amount of highly reliable pesticide registration and government test results.

[0148] In this embodiment, the evaluation model includes a first evaluation model and a second evaluation model. The machine learning unit 134 generates the first evaluation model by performing machine learning using first learning data in which the soil quality, topography, and climate of the test site are associated with the test report in the pesticide registration and the measured value or predicted value of the evaluation result of the test by the government. The second evaluation model is a model in which the measured value or predicted value is associated with the evaluation value. As a result, the pesticide management system S1 can perform machine learning using measured values ​​or predicted values ​​recorded in a test report, etc., even if the evaluation value is not recorded in the test report, etc., and can generate an evaluation model.

[0149] In this embodiment, the area information acquisition unit 132 (an example of a test area information acquisition unit) acquires weather information indicating the amount of rainfall on the test date or after the test. The machine learning unit 134 generates an evaluation model by performing machine learning based on the test area information indicating the soil quality, topography, climate, and rainfall of the test site and an evaluation value corresponding to at least two of the consumer risk, the exposure risk, and the ecological impact risk. The evaluation unit 3333 calculates the evaluation value based on the target area information indicating the soil quality, topography, climate, and rainfall of the target area and the evaluation model. This allows the pesticide management system S1 to calculate an evaluation value based on the amount of rainfall in the area. For example, if rain falls before the pesticide that has been sprayed and adheres to the crops dries, it may be washed away from the crops by the rain, reducing its efficacy. In addition, since the pesticide washed away by the rain flows into the environment, the evaluation value may be different from that in the case where the pesticide adheres to the crops. The pesticide management system S1 predicts an evaluation value based on the amount of rainfall in the area, and therefore can output an evaluation value with high accuracy. The amount of rainfall may be the amount of rainfall at the timing of the rainfall (the time of rainfall or the day of rainfall).

[0150] In addition, in this embodiment, the machine learning unit 134 generates a first evaluation model by performing machine learning using additional learning data that associates regional related information about the spraying location (an example of an application location) where the proposed pesticide was sprayed with the measured values ​​or predicted values ​​of the detection results after the pesticide was sprayed. This allows the pesticide management system S1 to perform machine learning using a larger amount of learning data, which may enable calculation of evaluation values ​​that are more adapted to the region.

[0151] In this embodiment, the exposure monitor 544 detects the presence of a person while spraying the pesticide. The evaluation unit 3333 calculates an evaluation value based on the detection result of the presence of a person. As a result, the pesticide management system S1 may be able to improve the accuracy of the evaluation value, for example, regarding the exposure risk evaluation. When the pesticide management system S1 uses an automatically operated drone 52, the pesticide management system S1 can reduce the worker exposure risk, and the effect of this can be reflected in the evaluation value.

[0152] The inventory management unit 335 manages the pesticides and the amounts of pesticides in containers T1 that can be loaded onto each base 51 and drone 52 (an example of pesticide spraying equipment) for each container T1. The inventory management unit 335 selects a container T1 based on a spraying request indicating the pesticide to be sprayed and the amount to be sprayed, and places an order for each container T1 filled with the pesticide to be sprayed if the amount to be sprayed is insufficient. The drone 52 is equipped with the container T1 filled with the pesticide to be sprayed, and sprays the pesticide. This allows the pesticide management system S1 to manage pesticides on a container T1 basis.

[0153] When ordering the container T1 filled with the pesticide to be sprayed, the inventory management unit 335 may order the container T1 in which the remaining amount is the same as or slightly different from the amount to be sprayed. This allows the pesticide management system S1 to order the pesticide required for spraying, and prevents the evaluation value from decreasing (risk increasing) due to the application or disposal of the remaining pesticide. The output section 213 may display information indicating each of the consumer risk, the exposure risk, and the ecological impact risk.

[0154] Also, the server 3 may automatically select one application plan from a plurality of application plans as the implementation application plan. The server 3 (for example, the spraying adjustment unit 334) selects the application plan with the highest evaluation value as the implementation application plan. In this case, the drone 52 flies and sprays the pesticide based on the implementation application plan. That is, the drone 52 (for example, the application unit 54) may control the flight and the spraying of the pesticide based on one or more evaluation values ​​according to at least two of the consumer risk, the exposure risk, and the ecological impact risk. Here, the server 3 or the drone 52 may notify or display on the user terminal 2 to obtain the user's permission for the flight and the spraying of the pesticide according to the selected implementation application plan or the implementation application plan. When the user's permission is obtained, the server 3 or the drone 52 may fly and spray the pesticide according to the selected implementation application plan or the implementation application plan. In other words, the drone 52 may control flight and spraying of pesticides based on one or more assessment values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk, and based on the user's permission.

[0155] [Note] (1) A pesticide management system that proposes the application of pesticides, comprising: a model memory unit that stores an evaluation model in which, for each pesticide and its application mode, regional information specific to the region is associated with an evaluation value corresponding to the risk of applying the pesticide in that region; an evaluation unit that calculates the evaluation value in a target region for each pesticide and its application mode based on target region information, which is the regional information of the target region, and the evaluation model; and a proposal unit that proposes the application of the pesticide based on the calculated evaluation value, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk.

[0156] (2) A pesticide management system as described in (1), comprising: a region-related information storage unit that stores region-related information associated with region information representing a location, soil type, topography, and climate; a pesticide information storage unit that stores pesticide-related information in which a target crop of a disease, disease information and damage scale representing a target crop of a disease, a disease name or a cause of the disease, and a pesticide to be sprayed on the target crop and an application mode indicating the amount of the pesticide to be sprayed; an input unit that inputs a pesticide spraying location, a target crop, disease information, and damage scale; a target region information acquisition unit that generates the target region information representing the soil type, topography, and climate of the occurrence location based on the spraying location and the region-related information; a pesticide selection unit that selects the pesticide and the application mode based on the target crop, disease information, damage scale, and pesticide-related information; and the evaluation unit that calculates the evaluation value at the occurrence location based on the generated target region information and the evaluation model for each of the selected pesticides and the application amounts of the pesticides indicated by the application modes.

[0157] (3) A pesticide management system as described in (2), comprising: a test information extraction unit that extracts the test location where the test was conducted, the consumer risk, the exposure risk, and the ecological impact risk for each pesticide and the application amount of the pesticide in the test from the test report in the pesticide registration and the pesticide test information indicating the evaluation results of the test by the government; a test area information acquisition unit that generates test area information representing the soil type, topography, and climate of the test location based on the test location and the area-related information; and a machine learning unit that generates the evaluation model by performing machine learning based on the test area information and the evaluation value corresponding to at least two of the consumer risk, the exposure risk, or the ecological impact risk.

[0158] (4) The pesticide management system described in (3), wherein the evaluation model includes a first evaluation model and a second evaluation model, and the machine learning unit generates a first evaluation model by performing machine learning using first learning data in which the soil type, topography, and climate of the test site are associated with the test report in the pesticide registration and the measured values ​​or predicted values ​​of the evaluation results of the test by the administrative side, and generates a second evaluation model using second learning data in which the measured values ​​or predicted values ​​are associated with the evaluation values.

[0159] (5) The pesticide management system described in (4), wherein the machine learning unit generates the first evaluation model by performing machine learning using additional learning data that associates regional information about the application location where the proposed pesticide was applied with measured or predicted values ​​of the detection results after spraying the pesticide.

[0160] (6) The pesticide management system described in (3), wherein the test area information acquisition unit acquires weather information indicating the amount of rainfall on the test date or after the test, the machine learning unit generates the evaluation model by performing machine learning based on test area information indicating the soil type, topography, climate, and rainfall of the test location and the evaluation value corresponding to at least two of the consumer risk, the exposure risk, or the ecological impact risk, and the evaluation unit calculates the evaluation value based on target area information indicating the soil type, topography, climate, and rainfall of the target area and the evaluation model.

[0161] (7) The pesticide management system described in (1) comprising an output unit that displays information indicating each of the consumer risk, the exposure risk, and the ecological impact risk.

[0162] (8) A pesticide management system as described in (1) above, wherein the exposure monitoring unit detects the presence of humans while the pesticide is being sprayed, and the evaluation unit recalculates the evaluation value based on the detection result of the presence of humans.

[0163] (9) A pesticide management system as described in (1), comprising: a pesticide management unit that manages the pesticides and their volumes injected into containers that can be mounted on a pesticide spraying device for each container; a container selection unit that selects the container based on a spraying request indicating the pesticide to be sprayed and the amount to be sprayed; and a pesticide spraying device that is mounted with the selected container and sprays the pesticide.

[0164] (10) A pesticide management method for proposing the application of a pesticide, comprising: storing an evaluation model in which, for each pesticide and its mode of application, region-specific regional information and an evaluation value corresponding to the risk of applying the pesticide in the region are associated with each other; calculating the evaluation value in a target region for each pesticide and its mode of application based on target region information, which is the regional information of the target region, and the evaluation model; and proposing the application of the pesticide based on the calculated evaluation value, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk.

[0165] (11) A pesticide management program that causes a computer of a pesticide management system that proposes the application of pesticides to execute a reading procedure for reading out an evaluation model in which, for each pesticide and its application mode, region-specific regional information and an evaluation value corresponding to the risk of applying the pesticide in that region are associated with each other, a risk calculation procedure for calculating the evaluation value in a target region for each pesticide and its application mode based on target region information, which is the regional information of the target region, and the evaluation model, and a proposal procedure for proposing the application of the pesticide based on the calculated evaluation value, wherein the evaluation value is one or more evaluation values ​​corresponding to at least two of consumer risk, exposure risk, and ecological impact risk.

[0166] FIG. 28 is an explanatory diagram illustrating the hardware configuration of each device according to this embodiment. Each device is a learning server 1, a user terminal 2, a management server 3, or an external server 4. Each device is configured to include an input / output module I, a storage module M, and a control module P. The input / output module I is realized by including a part or all of a communication module H11, a connection module H12, a pointing device H21, a keyboard H22, a display H23, a button H3, a microphone H41, a speaker H42, a camera H51, or a sensor H52. The storage module M is realized by including a drive H7. The storage module M may further be configured by including a part or all of a memory H8. The control module P is realized by including a memory H8 and a processor H9. These hardware components are connected to each other so as to be able to communicate with each other via a bus, and are supplied with power from a power source H6.

[0167] The connection module H12 is a digital input / output port such as a USB (Universal Seriul Bus). In the case of a portable device, the pointing device H21, the keyboard H22, and the display H23 are touch panels. The sensor H52 is an acceleration sensor, a gyro sensor, a GPS receiving module, a proximity sensor, or the like. The power source H6 is a power supply unit that supplies electricity required to operate each device. In the case of a portable device, the power source H6 is a battery. The drive H7 is an auxiliary storage medium such as a hard disk drive or a solid state drive. The drive H7 may be a non-volatile memory such as an EEPROM or a flash memory, or a magneto-optical disk drive or a flexible disk drive. In addition, the drive H7 is not limited to a drive built into each device, but may be an external storage device connected to a connector of the IF module H12. The memory H8 is a main storage medium such as a random access memory. The memory H8 may be a cache memory. The memory H8 stores instructions when the instructions are executed by one or more processors H9. The processor H9 is a CPU (Central Processing Unit). The processor H9 may be an MPU (Microprocessing Unit) or a GPU (Graphics Processing Unit). The processor H9 reads out programs and various data from the drive H7 via the memory H8 and performs calculations to execute instructions stored in one or more memories H8.

[0168] The input / output module I realizes an input / output unit I1, I2, or I3. The memory module M realizes a memory unit M1, M2, or M3. The control module P realizes a control unit P1, P2, or P3. In this specification, etc., the descriptions of the learning server 1, the user terminal 2, or the management server 3 may be replaced with descriptions of the control units P1, P2, or P3, respectively, or with descriptions of one or more processors. The learning server 1, the user terminal 2, or the management server 3 are devices that cause one or more processors to execute instructions using one or more memories that store instructions when the instructions are executed by the one or more processors.

[0169] In addition, a part of the learning server 1, the user terminal 2, the management server 3, or the external server 4 in the above-mentioned embodiment may be realized by a computer. In that case, a program for realizing this control function may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer system and executed to realize the control function. In addition, the "computer system" referred to here is a computer system built into the learning server 1, the user terminal 2, the management server 3, or the external server 4, and includes hardware such as an OS and peripheral devices. In addition, the "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into a computer system. Furthermore, the "computer-readable recording medium" may also include a device that dynamically holds a program for a short period of time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a device that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. In addition, the above-mentioned program may be a device for realizing a part of the above-mentioned function, and may further be a device that can realize the above-mentioned function in combination with a program already recorded in the computer system. In addition, some or all of the learning server 1, user terminal 2, management server 3, and external server 4 in the above-mentioned embodiments may be realized as integrated circuits such as LSI (Large Scale Integration). Each functional block of the learning server 1, user terminal 2, management server 3, and external server 4 may be individually processed, or some or all of them may be integrated into a processor. The integrated circuit method is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. Furthermore, if an integrated circuit technology that can replace LSI appears due to advances in semiconductor technology, an integrated circuit based on that technology may be used.

[0170] Although one embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes, etc. are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0171] S1...Pesticide management system, P...Administrative server, 1...Learning server, 2...User terminal, 3...Management server, 4...External server, 5...Spraying equipment, 51...Base, 52...Drone, 53...Monitoring device, I1, I2, I3, I5...Interface, M1, M2, M3, M5...Memory, P1, P2, P3, P5...Processor, D5...Drive unit, 111...Communication unit, 121...Test information storage unit, 122...Regional related information storage unit, 123...Learning data storage unit , 124... Additional learning data storage unit, 125... Learned model storage unit, 131... Test information extraction unit, 132... Regional information acquisition unit, 133... Learning data generation unit, 134... Machine learning unit, 135... Setting unit, T1... Container, 211... Communication unit, 212... Input unit, 213... Output unit, 214... Imaging unit, 215... Sensor unit, 221... Operation storage unit, 222... Application storage unit, 223... Data storage unit, 231... Operation unit, 24... Application unit, 25...functional section, 241...disease section, 242...spraying evaluation section, 243...proposal section, 244...inventory confirmation section, 245...spraying confirmation section, 311...communication section, 321...user information storage section, 322...model storage section, 323...region-related information storage section, 324...inventory information storage section, 325...equipment information storage section, 326...control history storage section, 331...user management section, 332...disease management section, 333...evaluation management section, 334...spraying adjustment section, 335...inventory management section, 336 ···Spraying application unit, 337···Spraying management unit, 511···Communication unit, 512···Input unit, 513···Output unit, 514···Imaging unit, 515···Sensor unit, 521···Spraying reservation memory unit, 522···Spraying status memory unit, 523···History memory unit, 531···Flight control unit, 54···Application unit, 55···Function unit, 541···Onboard setting unit, 542···Flight path unit, 543···Spraying control unit, 544···Exposure monitoring unit, 545···Spraying report unit, 561···Flight drive unit, 562···Spraying drive unit< / url:>

Claims

1. A pesticide management system that proposes the application of pesticides, a model storage unit that stores an evaluation model in which regional information specific to a region is associated with an evaluation value according to the risk of applying the pesticide in the region for each pesticide and its application mode; An evaluation unit that calculates the evaluation value in the target area for each pesticide and its application mode based on target area information, which is the regional information of the target area, and the evaluation model; A suggestion unit that suggests application of a pesticide based on the calculated evaluation value; Equipped with The evaluation value is one or more evaluation values ​​according to at least two of consumer risk, exposure risk, and ecological impact risk. Pesticide management system.

2. a region-related information storage unit that stores region-related information associated with region information representing a location, soil type, topography, and climate; a pesticide information storage unit that stores pesticide-related information in which disease information and damage scale indicating a target crop, a disease name, or a cause of a disease, and a pesticide to be sprayed on the target crop and an application mode indicating the application amount of the pesticide are previously associated with each other; an input section for inputting the location of pesticide application, the target crop, disease information, and the scale of damage; a target area information acquisition unit that generates the target area information representing the soil type, topography, and climate of the spraying location based on the spraying location and the area-related information; A pesticide selection unit that selects the pesticide and the application mode based on the target crop, disease information, damage scale, and pesticide-related information; The evaluation unit calculates the evaluation value at the spraying location based on the generated target area information and the evaluation model for each of the selected pesticides and the application amounts of the pesticides indicated by the application modes. The pesticide management system of claim 1.

3. a test information extraction unit that extracts information on the test site where the test was conducted, consumer risk, exposure risk, and ecological impact risk for each pesticide and application amount of the pesticide from the test report for pesticide registration and the pesticide test information showing the evaluation result of the test by the government; a test area information acquisition unit that generates test area information representing soil quality, topography, and climate of the test area based on the test area and the area-related information; a machine learning unit that generates the assessment model by performing machine learning based on the test area information and the assessment value corresponding to at least two of the consumer risk, the exposure risk, and the ecological impact risk; The pesticide management system according to claim 2 , comprising:

4. the evaluation models include a first evaluation model and a second evaluation model; The machine learning unit is A first evaluation model is generated by performing machine learning using first learning data in which the soil, topography, and climate of the test site are associated with the test report in the pesticide registration and the measured or predicted values ​​of the evaluation results of the test by the government, The second evaluation model is a model in which the measurement value or the predicted value and the evaluation value are associated with each other. The pesticide management system of claim 3.

5. The machine learning unit performs machine learning using additional learning data in which regional information on the application site where the proposed pesticide was applied is associated with measured values ​​or predicted values ​​of the detection results after the pesticide was sprayed, thereby generating the first evaluation model. The pesticide management system according to claim 4 , comprising:

6. The test area information acquisition unit acquires weather information indicating the amount of rainfall on the test day or after the test, the machine learning unit generates the assessment model by performing machine learning based on test area information representing soil quality, topography, climate, and rainfall of a test location and the assessment value corresponding to at least two of the consumer risk, the exposure risk, and the ecological impact risk; The evaluation unit calculates the evaluation value based on target area information representing soil quality, topography, climate, and rainfall of the target area and the evaluation model. The pesticide management system of claim 3.

7. an output unit that displays information indicating each of the consumer risk, the exposure risk, and the ecological impact risk; The pesticide management system of claim 1 , comprising:

8. an exposure monitoring unit that detects the presence of a person when the pesticide is being sprayed; The evaluation unit recalculates the evaluation value based on a result of detecting the presence of a person. The pesticide management system of claim 1.

9. an inventory management department that manages the pesticides and their quantities in containers that can be loaded on each base and each pesticide spraying device, selects containers based on spraying requests that indicate the pesticides to be sprayed and the amount to be sprayed, and places an order for each container filled with the pesticide to be sprayed if the amount is insufficient; a pesticide spraying device that is equipped with a container filled with the pesticide to be sprayed and sprays the pesticide; The pesticide management system of claim 1 , comprising:

10. A pesticide management method for proposing application of a pesticide, comprising: storing an evaluation model in which regional information specific to a region is associated with an evaluation value according to the risk of applying the pesticide in the region for each pesticide and its application mode; Calculating the evaluation value in the target area for each pesticide and its application mode based on target area information, which is the regional information of the target area, and the evaluation model; Proposing application of a pesticide based on the calculated evaluation value; The evaluation value is one or more evaluation values ​​according to at least two of consumer risk, exposure risk, and ecological impact risk. Pesticide management methods.

11. The computer in the pesticide management system suggests the application of pesticides. a reading step of reading an assessment model in which regional information specific to a region is associated with an assessment value according to the risk of application of the pesticide in the region for each pesticide and its application mode; A risk calculation step of calculating the evaluation value in the target area for each pesticide and its application mode based on target area information, which is the regional information of the target area, and the evaluation model; A proposal procedure for proposing application of a pesticide based on the calculated evaluation value; Run the command, The evaluation value is one or more evaluation values ​​according to at least two of consumer risk, exposure risk, and ecological impact risk. Pesticide management programs.

Images