WO2023150851A1 - Procédé quantitatif et qualitatif mis en oeuvre par ordinateur pour recommander des traitements dans la lutte chimique contre les mauvaises herbes - Google Patents

Procédé quantitatif et qualitatif mis en oeuvre par ordinateur pour recommander des traitements dans la lutte chimique contre les mauvaises herbes Download PDF

Info

Publication number
WO2023150851A1
WO2023150851A1 PCT/BR2023/050047 BR2023050047W WO2023150851A1 WO 2023150851 A1 WO2023150851 A1 WO 2023150851A1 BR 2023050047 W BR2023050047 W BR 2023050047W WO 2023150851 A1 WO2023150851 A1 WO 2023150851A1
Authority
WO
WIPO (PCT)
Prior art keywords
database
herbicide
herbicides
emergence
weed
Prior art date
Application number
PCT/BR2023/050047
Other languages
English (en)
Portuguese (pt)
Inventor
Pedro TEIXEIRA SPINOLA
Luiz Henrique FRANCO DE CAMPOS
Lucas FALCONE DE REZENDE
Original Assignee
Teixeira Spinola Pedro
Franco De Campos Luiz Henrique
Falcone De Rezende Lucas
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Teixeira Spinola Pedro, Franco De Campos Luiz Henrique, Falcone De Rezende Lucas filed Critical Teixeira Spinola Pedro
Publication of WO2023150851A1 publication Critical patent/WO2023150851A1/fr

Links

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G13/00—Protecting plants
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01B—SOIL WORKING IN AGRICULTURE OR FORESTRY; PARTS, DETAILS, OR ACCESSORIES OF AGRICULTURAL MACHINES OR IMPLEMENTS, IN GENERAL
    • A01B79/00—Methods for working soil
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01C—PLANTING; SOWING; FERTILISING
    • A01C21/00—Methods of fertilising, sowing or planting

Definitions

  • This patent describes a method that is capable of suggesting all possible treatments, composed of herbicides alone or in mixture, and their respective dose(s), for effective and efficient control of( s) weed(s) in a given area of any size and type of cultivation, in any modality of intervention/operation.
  • the method takes into account the specific characteristics of the area, modality, weeds, and herbicides, including price, and allows the user, from the analysis of the list of suggestions, to select precisely which treatment he wants to apply or plan to apply.
  • the quantitative and qualitative method is implemented on a computer in the form of an algorithm that performs calculations and verifications based on parameters stored in a database, and according to user input, which specifies the characteristics of the operation, the area and the infesting flora of weeds (matology), in order to quickly result in a list of suggestions of effective treatments (effective and efficient) for that specific operation-area, which can be ordered by total cost of treatment or according to other metrics and treatment information collected during the execution of the algorithm.
  • the sequence of instructions automatically executed by the algorithm could, in theory, be executed manually by one person or a group of people, but the necessary amount of calculations and/or verifications, in some cases in the tens of millions, among other factors, such as the risk of human error, make this possibility unfeasible in practice.
  • a specific non-linear optimization procedure is used, which is a machine learning technology that demands very high computational resources, which reinforces the need for computer implementation.
  • Said database contains all the agronomic parameters relevant data, including the price per liter or kilogram of each herbicide, which can be configured and reconfigured in order to influence how the method works and, therefore, the results produced by it. This characteristic also allows said parameters to be continuously calibrated by the user in order to make the results produced more and more accurate.
  • some fixed parameters are also relevant, such as characteristics of the area that may differ from other areas of the same producer, including organic matter content and soil texture; presence of neighboring crops sensitive to certain herbicides; whether the cultivated variety is sensitive to herbicides or not; whether there are plans for another crop sensitive to certain herbicides in that area in the near future; if there was soil disturbance; whether there is neighboring human settlement; if there is straw present in the area and what is its management regime; recommended dose of herbicides in each case; effectiveness and efficiency of control of each herbicide for each weed; residual time of active surface and subsurface treatment; and analysis of the interaction and compatibility between different herbicides when used together.
  • Other variables relevant to an effective recommendation include the ability of some weeds to germinate in the subsurface; the ability of the weed to absorb herbicides that are on the subsurface when it is on the surface; the ability of herbicides to manage to control a plant when it is already germinated, but not emerged; among others.
  • the process of recommending and purchasing herbicides in these large producers generally presents the following steps, in order: a) The specialist is responsible for recommending the herbicides for operations in each area, for the all year round, and for all areas. b) The specialist needs to consult the specific conditions of each area, such as matology, location, soil, expected stage of the crop, among others. These consultations are normally carried out in spreadsheets, in notes in notebooks, or directly from an employee who knows the areas in person, and it is common that there is unavailable or incorrect information; c) The specialist will then define the operations that will be necessary throughout the year for each area, and a single treatment to be applied for each of these operations, in order to effect satisfactory control of the weeds present there.
  • an external specialist consultant is called in at this stage to provide assistance to the internal specialist; d) When the report containing the recommendations for each operation in each area is completed, it is forwarded to the Purchasing sector (also called Supplies or Commercial), which in turn compiles the data determining the total purchase quantity of each of the herbicides ; e) The commercial manager proceeds with assessments and common commercial procedures such as possible suppliers, contracts, price and quantity negotiations, among others; f) During the previous stage, it is common for the commercial manager to ask the specialist in cultural practices to make changes to the herbicide recommendations in order to change the final composition of the budget in one way or another, which consumes time for this specialist that he may not have it available, since to satisfy the commercial manager he will have to change some or several of his recommendations.
  • herbicides are marketed accompanied by a package insert with information on: which weeds they control, in which modalities and times they can be used, at which stage there is selectivity, and the amplitude of the dose that can be used.
  • the Crop Decisor application entered the market, in which weed species in the area are inserted, and the application suggests a single combination of herbicides, using information from the product leaflet (without customization) and without including the prices of the product, very similar to the previously mentioned spreadsheet-based systems, but offering greater usability to users.
  • WO20191 13500A1 discloses methods of comparing information on commercial herbicides and commercial seed varieties, and the commercial applications of these methods.
  • the user plans his crop by entering the herbicides he plans to use, which seed varieties, the history of his area and neighboring crops, and the system returns the probabilities of problems, conflict alerts in herbicide resistance , and whether or not to move forward.
  • the user inserts some herbicide that he intends to use and the system calculates whether or not there is a risk of carrying out this operation.
  • an embedded technology alerts the user that the selected herbicide may or may not cause a problem in the area whose herbicide application would be starting at that moment.
  • n s . CN102026545A discloses a method of controlling unwanted vegetation in a growing area. This method consists of, after applying a post-emergence herbicide to the site that also has a residual effect on the soil (pre-emergence effect) of some specific group listed, subsequently, the same or different post-emergence herbicides may or may not be applied to the site. emergent from some specific group(s) listed in another post-emergent application. This method brings about a reduction in the amount of post-emergence herbicide needed to control unwanted vegetation in the second application which can be achieved by the first application being also a pre-emergence treatment.
  • the second application may or may not include the same or another post-emergent herbicide(s) from some specific group(s) listed for the site, providing a post-emergence treatment. -emerging with a reduced amount of the herbicide or different post-emergent herbicides. This is achieved by emergence being delayed by pre-emergence treatment control and the resulting post-emergence being of younger stages and can be controlled with less product in the post-emergence treatment used after the first one.
  • This method substantially differs from the method described in the present patent.
  • the document by n s . CA2882940A1 reveals an agricultural input indication system. Methods, apparatus and computer program products are provided to produce targeted indications of agricultural inputs based on a given context of geolocated use.
  • Methods include receiving one or more nominations in the context of geolocated use, determining one or more suggested agricultural inputs based on the context of use, and having one or more suggested agricultural inputs provided.
  • a plurality of usage scenarios may be presented for selection, each of the usage scenarios being associated with one or more further indications in the context of geolocated usage.
  • the probabilities of reaching the target and the minimum acceptable yields can be determined and presented together with the climatic scenarios.
  • Document No. US20160308954 discloses a system and method for processing and managing cloud-based agricultural data for data integration with agricultural operations based on geolocation; the system of the present patent, however, is independent of geolocation and does not use geolocated data to store any information for later consultation.
  • the prior system takes agriculture-related data associated with a given geographic area and transforms the received data into an analysis-ready format; the present patent is not a tool that facilitates the interpretation of data, but a tool that receives data and delivers answers.
  • the prior system processes the data received by means of one or more algorithms for determining at least one operation to be carried out within the determined geographic area; already in the present patent, it is not determined which operation to perform, but which input(s) to use in a given operation to be able to solve a given problem.
  • the previous system generates a set of instructions for executing at least one operation within a given geographic area based on geolocation, where the instructions are encoded for direct use by a controller of a given type of agricultural equipment; and transmits the instructions through a wireless communication channel to the controller, where the instructions cause the controller to direct the operation of the agricultural equipment to perform at least one operation within the geographical area determined according to the geolocation in an automated way - the system appears to be for managing plots and agricultural operations, being distinct from the patent proposal.
  • the document does not specifically mention the feature of calculating interactions between characteristics of at least one area of user input and pre-programmed data.
  • paragraphs 78 and 193 of this same document US20160308954 it can be understood that the prior art reveals a computer-implemented method for indicating herbicides and dosages in order to make treatment for different crops in the control of weeds.
  • paragraph 78 refers to a method that consults a package of inputs, including herbicides, to verify that the product is indicated to solve the indicated problem.
  • this method here cannot combine inputs and can only indicate a solution to a simple problem (a single weed or at most multiple weeds, but only those that are controlled by one and the same herbicide). .
  • the present patent indicates all the solutions for complex problems (mixtures of herbicides and multiple weeds).
  • paragraph 73 addresses about hosting the database in the cloud; paragraph 74 on organizing this data in a user-usable manner; paragraphs 92 and 100 on how to handle this database; paragraph 170 about how the areas of each farm will be registered and paragraph 268 about the pre-programmed recommendations in the system and how the data and connected equipment will use it.
  • This system looks, in fact, an ERP (Integrated Business Management System), which is not the case of the present patent.
  • paragraph 6 demonstrates the primary difference between the prior art and the present patent: the prior art system serves to register solutions to problems: “The system also includes a prescription generation module defined to generate a set of instructions for execution of the at least one operation within the given geographic area as a function of geolocation 11 .
  • the solutions are calculated directly from the particulars of the problem; in the anteriority, if you have “A”, use “B”; in the present patent if you have “A”, through calculations, it is recommended to use “B”, “C”, “D”, “E”, “F” or “G”, sorted by price.
  • the database of the present patent does not register solutions, but rather finds solutions to registered problems - that is, it dynamically registers problems and searches for answers at that moment.
  • the method described in the present patent does not depend on any type of pre-registration of ready-made solutions.
  • It comprises a plurality of modular components for receiving and processing data relating to the agricultural production of commodities by an agricultural producer and for centralizing and storing the received and/or processed data in a single cloud-based database.
  • the producer may provide one or more third-party vendors and/or service providers with authorized but restricted access to selected components of its agricultural business management system and cloud-based database so that, together, the producer , suppliers and service providers can effectively plan and cost-effectively manage the delivery of products and services during an agricultural production cycle and the sale of harvested agricultural commodities.
  • Separate modular components can even be provided for inputs exemplified by agronomy data, agricultural production input data, and tracking growth and crop performance. It is a logistics system for goods and inputs specific to the agro sector, but which has no proximity to the proposal of the present patent.
  • a central advantage of the invention is the fact that it can be applied to aid in the recommendation of herbicides for any type of intervention; for any agricultural crop including pastures; to control any weeds, in any quantity and at any stage of emergence; and using only the necessary herbicides, at the required dose, in isolation or in a mixture. In this way, it allows an effective and efficient chemical control, being applied to most of the real situations encountered by farmers in their plots.
  • Another central advantage lies in the fact that the method guarantees that all calculations that produce the suggested treatments for a given operation-area not only consider all variables of interest to the user, but also never include calculation errors, as they are performed by computer. In addition, unlike other indicated methods and/or heuristic models, this method does not depend on pre-registration of treatments, and checks all calculations in relation to all configured parameters, every time the algorithm is executed.
  • the present method allows thousands or millions of calculations to be performed in an almost instantaneous time interval to produce recommendation suggestions - while in the conventional recommendation, to produce a single recommendation for an operation-area , it takes considerable time to evaluate all the necessary variables and ensure the availability of the desired herbicides in stock/market. Due to the recommendation speed, the ability to generate several effective assertive options, among other essential characteristics of the method, it is clear that it can be used both to make instant recommendations and to plan the recommendation for an operation-area.
  • Another essential advantage of the method is being able to produce a complete list with all the available options of effective chemical treatments for an operation-area, according to configured parameters, while other methods and/or existing heuristic models very commonly only offer one treatment suggestion. This allows for a significant increase in the flexibility of the recommendation process.
  • the present method allows technicians specialized in recommendations, who were previously forced by several factors to prioritize the effectiveness of control of the recommendation to the detriment of its efficiency, now have access to a method that allows them to optimize to the recommendations reaching the best cost-benefit possible, reducing the production cost of the ton of the crop in question and considerably increasing the profitability of the crop.
  • a primary advantage provided by the present method within the scope of this invention is that it can be easily replicated for different operations-areas, being able to promote its advantages for each of them, and allowing agricultural productions to have an optimized management of weed control in all their areas of cultivation. Additionally, the method allows the farmer to further subdivide his cultivation areas, obtaining even greater precision and effectiveness, as well as a possible reduction in costs.
  • the method can be implemented both in a local environment and in the cloud, including through a digital platform. It can therefore be accessed via the internet, mobile devices and others such as precision agriculture equipment for the application of different herbicides alternately in the same operation area, according to the application map, such as those of Raven Industries, INC® . Thus, making access to good herbicide recommendations more accessible.
  • the database that integrates the present method can be customized and calibrated, which allows advantages such as the addition or exclusion of a herbicide or weed, for example, or the effective control of even strains/ subspecies resistant to certain herbicides, provided herbicides are available for their control.
  • the feedback speed of the system represented by the method is greater: in other words, it is possible to quickly understand how a certain change in the parameterization produces or fails to produce certain result(s)( s), or the reverse.
  • This characteristic makes evident the benefit that the present method provides for learning related to the recommendation of herbicides, allowing the trained operator to learn and develop continuously and faster than the previously available methods, passively, while operating the tool.
  • this method can easily be used with fictitious herbicides and/or weeds, for didactic reasons.
  • Another benefit of this advantage is that it facilitates method operators in refining the parameters, thus enabling increasingly accurate and efficient results.
  • Another essential advantage of the present method since it allows for other innovations mentioned here, consists of the fact that it allows the specific storage of the results produced by the algorithm for an area-operation, or for multiple areas-operations - in the case of the method is thus replicated; and the data produced by the algorithm.
  • this feature allows you to perform all sorts of isolated or aggregated analysis, in any form, regarding any stored metrics. This type of business intelligence can lead to several improvements in metrics and/or user processes.
  • Another advantage of the invention is that it allows automatic updating of the recommendation suggestion. For example: the user has already used the method to generate a list of suggested treatments for a given operation-area; he has already selected one of the treatments as chosen; if the prices of some herbicides change on the market, he can then change them in the parameterization; results are automatically updated; and now the previously generated list may have changed its order, therefore there may be a more efficient option than the previously selected one.
  • Another advantage given to users of this method, as it is implemented on a computer, in situations where the control determined by the user is above the current capabilities of the herbicides, according to the relevant parameterizations, consists of applying a heuristic that varies the input in search of valid results that may indicate a possible strategy for appropriate weed control (do two pre-emergence applications with half of the residue each, program a chemical weeding afterwards, among other possibilities).
  • Another advantage consists of being able to feed back based on configurable parameters.
  • an area treated with a pre-emergence recommendation generated by the method has ended, it can be deduced that either the infestation of the area was underestimated (if there is only one area with this profile) or the parameters of the herbicides in question were overestimated (if all areas that received this treatment with similar infestation had this behavior), or there was an abnormal climatic condition that modified the expected behavior of the herbicides (if all areas with similar infestation at a given time had this behavior).
  • the analysis of the result from the perspective of the present method allows feed back the base of configurable parameters with these results and refine it, improving the user's knowledge on the subject and mainly the future result of weed control, in addition to allowing monitoring the evolution of the seed bank of their areas.
  • Another advantage conferred by the present method, especially to large producers, and arising from other previously explained advantages, is that facilitates the internal standardization of processes related to weed control, which these groups often have as a strategic objective (to meet compliance standards, for example) and can be fragmented among different people and/or production units belonging to the same organization.
  • Another essential advantage of the method is the inclusion of the price dimension of each herbicide according to the actual conditions of acquisition by the user, who can configure these parameters in the database - functionality that is not foreseen in methods and/or heuristics with similar objectives to the one mentioned above.
  • This feature is not only one of the main factors that contribute to increasing the efficiency of recommendations, but also “unlocks”, in turn, several other advantages, some of which will be explained below.
  • This method can be implemented on a computer in several different ways, with many different objectives, being the basis for different applications.
  • this enables implementations that guarantee the user greater usability than that offered by at least some of the previously mentioned priorities, as well as greater flexibility for customizing these implementations. It's just that said priorities are restricted to the specific implementation (for example, in Microsoft Excel), and the present method needs to be implemented in a computer but is not restricted to a single implementation.
  • the present method can be applied in order to enable the semi-automation of the herbicide budgeting process. Is that, in the conventional way in which this process is conducted, several possible missed opportunities and/or direct losses in the productivity of the production unit are caused.
  • the method allows these professionals to carry out various types of analyzes and comparative studies on price and positioning not only of their organization's products, but also of competing products. There are added advantages that the application of the method could provide for these organizations, such as the ability to carry out advanced market studies and/or identify opportunities in the segment.
  • Another advantage of this method is the fact that it enables and encourages a more responsible management of weed control in agricultural production in order to ensure animal, human and environmental safety.
  • the database as it is customizable, can include specific agronomic parameters that contribute in this sense, such as the one referring to human settlements neighboring the cultivation area.
  • Figure 1 represents a general flowchart of the steps of the computer-implemented method for recommending herbicides and dosages in order to compose treatments for effective weed control in a crop area;
  • Figure 2 represents a flowchart of the specific steps that integrate the quantitative and qualitative algorithm.
  • the method implemented on a computer or analogous device integrates an algorithm (ALG) of quantitative (calculations) and qualitative (verifications) instructions performed on parameters stored in a database (DB), which can be calibrated by the user (US1) or another, and on the input (INP) that the user fills in to characterize the operation-area for which he intends to make a recommendation.
  • AAG algorithm of quantitative (calculations) and qualitative (verifications) instructions performed on parameters stored in a database (DB), which can be calibrated by the user (US1) or another, and on the input (INP) that the user fills in to characterize the operation-area for which he intends to make a recommendation.
  • the algorithm is able to produce a list of all effective treatments (TRATS) for the target-area operation.
  • Said parameters refer to numerical (quantitative) price values for each herbicide, and those that represent the environment-weed-herbicide system of the crop in question, modeled according to the form proposed by Rezende (2021), co-inventor of this patent , and which allows the chemical control of any cultivated species, with adequate parameterization. They also include (qualitative) information and numerical values related to the cultivated genus, the herbicides that will be considered possible, the weeds that will be considered possible, the intervention modalities that will be considered possible, and some of the relationships between these elements as variables of interest. by cultivation and modality.
  • the operations/interventions are subdivided into different modalities, each of which determines, according to technical criteria, a different moment, application technology and purpose for carrying out the intervention. From the point of view of the method, what distinguishes different modalities are two elements: which parameters in the database will be considered for the calculation, and which herbicides can be used in this modality. Many crops share the same modalities with sugarcane, for example, but their exact functioning can differ: both in sugarcane and soybeans, for example, there is a modality called “post-planting”, but the parameterization as well as the variables of interest for the execution of the algorithm, stored in the database are different. The present method therefore makes it possible, through parameterization in the database (DB), to include/exclude and customize any type of intervention.
  • DB parameterization in the database
  • the present method can be used to generate treatment suggestions for any crop in which chemical weed control is practiced, and for any type of intervention.
  • herbicides and which weeds the algorithm specifically takes into account are themselves parameters contained in the database (DB)
  • this allows the method to be used also to provide suggestions for treatments containing any herbicides, and to control any weeds, as long as they are parameterized in the database.
  • the algorithm verifies and/ or calculates the interactions between this information and the parameters (DB), in order to result in a list (TRATS) containing all suggestions for possible effective treatments composed of herbicides registered in the database, together or separately, and their respective dosages, in addition to information such as total cost per hectare and other data (DT) on each of these treatments.
  • the data (DT) specifically concerns the specific information and metrics for each treatment (and for the herbicide(s) that compose(s) it), collected during the execution of each of the sequential steps of the algorithm (ALG) , and all possible information and metrics for each treatment are collected.
  • These data (DT) serve not only so that the user can order the list (TRATS) of treatments using them as a parameter, but also so that the user can, after producing the output (OUT), consult some specific data about some treatment (or herbicide(s)). In addition, they can be stored for later consultation (ARM), together with the list (TRATS).
  • step (ET7) it may happen that the dose of some herbicides in a given treatment is changed - if this occurs, the following data (DT) will be recorded for this treatment: a) What the treatment had some change in dose during stage (ET7) (information); b) Which herbicide or herbicides had dose modification (information); c) What was the previous dose and what was the new dose, or difference between them (metrics).
  • DT specific data
  • Other examples of said data (DT) may include, by way of example and not limitation: dose of a herbicide per hectare; price per liter of an herbicide; herbicide cost per hectare (price multiplied by dose); total treatment cost per hectare (addition of the individual costs of each cost per hectare of each herbicide that composes it); which herbicides within the treatment are effective to control pre-emergence, or post-emergence, or both; control efficiency for said weed of matology (11 ), etc.
  • the input itself (INP) is registered for each treatment, for example.
  • the method can be used both in the field, immediately, to generate treatment suggestions so that the user can select the best option (recommendation) for a given intervention of a given modality in a given area, and for prior weed control planning.
  • the user can change the ordering of the list of suggestions to use any treatment data, such as the cheapest per hectare to the most expensive.
  • the list in turn, can be stored in the database (ARM) or in another specific database.
  • the database (DB) parameters mentioned above include, by way of example and not limitation:
  • Characteristics of each specific operation area to be treated such as: soil texture (high, medium or low); soil organic matter content (high, medium or low); existence of a neighboring crop susceptible to herbicides; existence of neighboring settlement; preparation or not of the soil with turning; expected planting date; prediction of incorporation in the application; crop stage in the area (for sugarcane, for example: pre-emergence, spur, up to four leaves or over four leaves); sensitivity or not to herbicides of the crop variety in the area; variety closing speed; time of application (wet, semi-wet, semi-dry or dry); presence or not of straw and its handling; crop rotation forecast in the year following the application; knowledge of the weed population in the area, as well as population pressure of each individual plant; and, if post-emergence is present, its stage of development, and definition of the control standard(s) (in %);
  • P2 Type of crop (soybean, corn, wheat, sorghum, canola, pastures, coffee, orange, among others), intervention modalities (incorporated pre-planting, post-planting or cane-plant, sequential, desiccation, cane- punches for example, among others);
  • Mt treatment mortality of a given weed
  • n amount of herbicides in the treatment
  • Xk Dose of herbicide k
  • D50k Dose to kill 50% of weed individuals by herbicide k;
  • Ck Sigmoid curvature of survival of herbicide k
  • Equation (EQ2) for mortality of the combination of “n” herbicides in pre-emergence [071] Equation (EQ2) for mortality of the combination of “n” herbicides in pre-emergence:
  • Mt treatment mortality of a given weed
  • n amount of herbicides in the treatment
  • Mk Weed mortality to herbicide k
  • the steps of the method include the execution of a global non-linear optimization procedure using the differential evolution method to determine the optimal dose in the case of weeds in post-emergence.
  • This procedure includes a target equation (EQ3) and constraint equations (EQ1 m):
  • the objective equation has the function of minimizing the total cost of the treatment being processed, and it must obey all the restrictions imposed by the restriction equations, which are multiple (one for each weed and one for each herbicide) , with constraints “1 to m” similar to equation (EQ1 ), but used as constraint equations (EQ1 m) in an optimizer algorithm:
  • Ik retrieved from the database.
  • Step (ET1 ) - The user (US1 ) specifies in the input (INP) specific data about the operation-area (P1 ), according to the modality (P2), as described in Table 1 , in the example of sugarcane :
  • Table 1 Variables considered in the area by modality, for sugarcane for example.
  • the user can, therefore, determine for an area in which he plans to operate ratoon cane, as an example:
  • Viola string in pre-emergence at low pressure Brachiaria in pre-emergence and medium pressure
  • Step (ET3) Combination of all remaining herbicides, according to the previous step, in order to compose potential treatments, each one containing from one herbicide to the total number of remaining herbicides. Treatments composed of mixtures of herbicides that are incompatible with each other, according to the database (P4), are then discarded;
  • Step (ET4) If there are weeds in pre-emergence in the matology (11 ), the remaining treatments are tested against the minimum mortality defined by the standard of control of the infestation pressure of the seed bank parameterized in the database (P3) and the mortalities of each herbicide for each input weed (11 ) and their interactions in the database (P5), by calculating the pre-emergence efficiency of each treatment for each weed using equation (EQ2), zeroing out the straw mortality depending on its handling (I3), or using it according to the database (P3). If there are no weeds in pre-emergence in the matology (11 ), after this step, steps are skipped (ET5-ET8);
  • Step (ET5) If it is necessary to control weed seeds in the subsurface according to the input (I3), the resulting mortality is calculated again for each remaining treatment only with the herbicides that act in depth according to the base of data (P4) and their interactions according to the database (P5), for weed species of the matology (11 ) that present germination in depth, according to the database (P3), using the same equation (EQ2 ). Treatments that do not meet are discarded;
  • Step (ET6) The herbicide doses of the remaining treatments are then retrieved from the database (P5) according to the soil organic matter content and soil texture, modality and time in the input (INP) - the objective is to dose the herbicides. If in the previous step (ET5) with the equation (EQ2) the combination was validated, that is, it was verified that there is a possibility of it controlling the problem, now in the present step, the herbicide doses are adjusted accordingly to control the target problem satisfactorily.
  • Step (ET7) If there is incorporation in the input (13), the dose of each herbicide can be modified, according to the database (P5), in the case of a high volatility herbicide (P4), in the remaining treatments;
  • Step (ET8) If relevant according to the straw management of the input (13), the herbicide doses of the remaining combinations are adjusted by the algorithm (ALG) by a straw correction factor for each application time for each herbicide , according to parameterization in the database (P4);
  • Step (ET9) If there is germane in the area, information entered by the user in the input (11 ), and treatments have been defined to control pre-emergence, these validated treatments are qualitatively tested, where the herbicides and their doses are compared on the basis (P5) with the herbicides and the doses needed to control germane. If there is any herbicide that controls, the treatment is validated. If they do not control it, add herbicides or a herbicide capable of treating the germ in the area. If there are no treatments for pre-emergence, the herbicides that treat the germ in the area are used as validated treatments. If there is no germane, according to the input (11 ), it goes to the next step (ET10);
  • Step (ET 10) If there is the presence of weeds in post-emergence, information entered by the user in the input (11 ), those that are mechanically controlled are eliminated from the list of weeds to be controlled by the sequence of the algorithm (ALG) for the incorporation of herbicides, according to the database (P3), if incorporation is foreseen for the target area, information also entered by the user in the input (I3). If there are no remaining weeds in the list of weeds and stages to be controlled, after this step, steps are skipped (ET1 1 -ET13) and the remaining treatments are added to the register of usable treatments;
  • Step (ET1 1 ) - The equations (EQ1 m) are then assembled for the weeds in post-emergence, with the parameters pre-programmed by the user in the database (P5), for each herbicide of the treatments, in each weed and stadium inserted by the user in the input (11); Step (ET12) -
  • the herbicide doses for each treatment are calculated through non-linear global optimization using the differential evolution method, minimizing the treatment cost, sum of the dose of each herbicide multiplied by the price, according to the database .
  • Step (ET14) If no treatment has yet been registered in the output (OUT) to treat that operation-area, a heuristic is applied in these cases: one of the database variables (P1 ) is varied in the input (INP ), and the algorithm (ALG) is executed again from step (ET2). For example: the residual time of a certain weed can be reduced, which can lead to the algorithm being able to find effective results in this execution.
  • This heuristic is repeated several times, automatically by the algorithm, until a valid result is found by the algorithm, and then the output (OUT) is formed as described in the previous step, but warning the user (US1 ) that these are alternative results . In some cases, however, even after If the algorithm is executed hundreds of times, it is possible that the possible variations run out: the user is then informed that it was not possible to find valid results or alternatives for that recommendation.
  • Step (ET15) The total cost per hectare of each usable treatment is calculated according to the database (P4) and the doses of the treatments found, and the treatment is recorded in the output (OUT), which contains the list of cost-effective treatments (TRATS) to facilitate the user's decision on which to apply.
  • the herbicides used in the previous application data entered by the user in the input (I3), are marked in red, and those from the same family of the previous application, according to the database, in yellow, in case he wants to avoid repeating herbicides/ families in order to avoid creating resistance to the herbicides in that weed area/population.
  • the output (OUT) also comprises the specific data (DT) produced for each treatment approved for the operation-area, by the algorithm. Algorithm execution ends and the method flowchart continues. With these results in hand, the user can then proceed to analyze and recommend one of the treatments.
  • the algorithm could produce the following list, ordered by total cost:
  • Clomazone6OO 1.5 liters/hectare

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Animal Husbandry (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • Mining & Mineral Resources (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Agronomy & Crop Science (AREA)
  • Mathematical Analysis (AREA)
  • Environmental Sciences (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Operations Research (AREA)
  • Pure & Applied Mathematics (AREA)
  • Game Theory and Decision Science (AREA)
  • Toxicology (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Algebra (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Quality & Reliability (AREA)
  • Mechanical Engineering (AREA)
  • Soil Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Agricultural Chemicals And Associated Chemicals (AREA)

Abstract

La présente invention concerne un procédé permettant de suggérer tous les traitements possibles composés d'herbicides isolés ou mélangés, et leurs doses respectives pour la lutte efficace et efficiente contre les mauvaises herbes dans une zone déterminée d'une taille ou d'un type de culture quelconque, dans n'importe quelle modalité d'intervention/opération. Le procédé quantitatif ou qualitatif est mis en œuvre PAR un ordinateur sous forme d'un algorithme qui effectue des calculs et des vérifications sur la base de paramètres stockés dans une base de données, et selon l'entrée de l'utilisateur, qui spécifie les caractéristiques de l'opération, de la zone et de la flore envahissante des mauvaises herbes de manière à déboucher, rapidement, sur une liste de suggestions de traitements efficaces pour cette opération-zone spécifique qui peut être constituée par prix ou selon d'autres critères qui comprennent des données spécifiques portant sur chaque traitement, générées pendant l'exécution de l'algorithme.
PCT/BR2023/050047 2022-02-10 2023-02-10 Procédé quantitatif et qualitatif mis en oeuvre par ordinateur pour recommander des traitements dans la lutte chimique contre les mauvaises herbes WO2023150851A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
BR1020220025410 2022-02-10
BR102022002541-0A BR102022002541A2 (pt) 2022-02-10 2022-02-10 Método quantitativo e qualitativo implementado em computador para indicação de herbicidas e dosagens a fim de formar tratamentos para culturas diversas no controle de plantas daninhas e indicação de aplicação resultante

Publications (1)

Publication Number Publication Date
WO2023150851A1 true WO2023150851A1 (fr) 2023-08-17

Family

ID=87563277

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/BR2023/050047 WO2023150851A1 (fr) 2022-02-10 2023-02-10 Procédé quantitatif et qualitatif mis en oeuvre par ordinateur pour recommander des traitements dans la lutte chimique contre les mauvaises herbes

Country Status (2)

Country Link
BR (1) BR102022002541A2 (fr)
WO (1) WO2023150851A1 (fr)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160308954A1 (en) * 2015-04-20 2016-10-20 Wilbur-Elllis Company Systems and Methods for Cloud-Based Agricultural Data Processing and Management
US9974226B2 (en) * 2014-04-21 2018-05-22 The Climate Corporation Generating an agriculture prescription
US10028426B2 (en) * 2015-04-17 2018-07-24 360 Yield Center, Llc Agronomic systems, methods and apparatuses
US20190043142A1 (en) * 2011-05-13 2019-02-07 The Climate Corporation Systems to prescribe and deliver fertilizer over agricultural fields and related methods
US20190050948A1 (en) * 2017-08-08 2019-02-14 Indigo Ag, Inc. Machine learning in agricultural planting, growing, and harvesting contexts
US20200265527A1 (en) * 2019-02-15 2020-08-20 Growers Edge Financial, Inc. Agronomic prescription product
US11069007B2 (en) * 2017-04-10 2021-07-20 Decisive Farming Corp. Crop management method and system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190043142A1 (en) * 2011-05-13 2019-02-07 The Climate Corporation Systems to prescribe and deliver fertilizer over agricultural fields and related methods
US9974226B2 (en) * 2014-04-21 2018-05-22 The Climate Corporation Generating an agriculture prescription
US10028426B2 (en) * 2015-04-17 2018-07-24 360 Yield Center, Llc Agronomic systems, methods and apparatuses
US20160308954A1 (en) * 2015-04-20 2016-10-20 Wilbur-Elllis Company Systems and Methods for Cloud-Based Agricultural Data Processing and Management
US11069007B2 (en) * 2017-04-10 2021-07-20 Decisive Farming Corp. Crop management method and system
US20190050948A1 (en) * 2017-08-08 2019-02-14 Indigo Ag, Inc. Machine learning in agricultural planting, growing, and harvesting contexts
US20200265527A1 (en) * 2019-02-15 2020-08-20 Growers Edge Financial, Inc. Agronomic prescription product

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
FALCONE DE REZENDE LUCAS: "OTIMIZAÇÃO DA ESCOLHA DE HERBICIDAS NA PRODUÇÃO DE CANA-DE-AÇÚCAR", DISSERTATION, SÃO PAULO/SP, 1 January 2021 (2021-01-01), SÃO PAULO/SP, XP093083876, Retrieved from the Internet <URL:https://bibliotecadigital.fgv.br/dspace/bitstream/handle/10438/31465/Disserta%C3%A7%C3%A3o%20Lucas%20Rezende.pdf?sequence=3> [retrieved on 20230920] *

Also Published As

Publication number Publication date
BR102022002541A2 (pt) 2023-12-26

Similar Documents

Publication Publication Date Title
US20230018607A1 (en) System and method of confirming standard compliance for at least one agricultural product
Sharma et al. Technology adoption and pest control strategies among UK cereal farmers: evidence from parametric and nonparametric count data models
Norton et al. Mathematical programming for economic analysis in agriculture
Winters et al. Designing impact evaluations for agricultural projects
Zhengfei et al. Integrating agronomic principles into production function specification: a dichotomy of growth inputs and facilitating inputs
Skevas et al. Reducing pesticide use and pesticide impact by productivity growth: the case of Dutch arable farming
Falconer et al. Using economic incentives for pesticide usage reductions: responsiveness to input taxation and agricultural systems
US20140278731A1 (en) System and Method for Agricultural Risk Management
Tekalign Forage seed systems in Ethiopia: A scoping study
Sperling When disaster strikes: a guide to assessing seed system security
Catarino et al. Fostering local crop-livestock integration via legume exchanges using an innovative integrated assessment and modelling approach based on the MAELIA platform
Hurt The green revolution in the Global South: Science, politics, and unintended consequences
Shi et al. An analysis of selectivity in the productivity evaluation of biotechnology: an application to corn
JP5479770B2 (ja) 農業経営支援プログラムおよび農業経営支援システム
Lacoste et al. Upgrading the RIM model for improved support of integrated weed management extension efforts in cropping systems
Nooghabi et al. Social, economic and environmental vulnerability: The case of wheat farmers in Northeast Iran
Kide Smallholder farmers’ adaptation strategies to climate change in Ethiopia: evidence from Adwa Woreda of Tigray region
Jones et al. The importance of seasonal variability and tactical responses to risk on estimating the economic benefits of integrated weed management
Galpin et al. Participatory farm management methods for agricultural research and extension: a training manual
WO2023150851A1 (fr) Procédé quantitatif et qualitatif mis en oeuvre par ordinateur pour recommander des traitements dans la lutte chimique contre les mauvaises herbes
BR102023002523A2 (pt) Método quantitativo e qualitativo implementado em computador para recomendação de tratamentos para o controle químico de plantas daninhas
Paul Jr et al. Harnessing digital innovations for climate action and market access: Opportunities and constraints in the CWANA region
Canney-Davison et al. Tools4SeedSystems: Working towards resilience through root, tuber and banana crops in humanitarian settings. Workshop Report.
Garrett et al. User guide to the seedHealth model as part of the integrated seed health approach. RTB User Guide
Mensah et al. Revisiting the synergy between no-till and roundup ready soybean technology

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23752172

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE