Device for the targeted application of a liquid to an agricultural field

The device and method address crop damage from herbicides by detecting and calculating potential wetting to control liquid application, ensuring precise and safe herbicide use, thus reducing phytotoxic reactions and maintaining crop health.

WO2026233830A1PCT designated stage Publication Date: 2026-11-12BAYER AG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/062470
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2026-04-30
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing agricultural spraying technologies often result in phytotoxic reactions and physiological shocks in crops due to herbicide application, particularly when non-selective herbicides are used, leading to potential yield losses.

Method used

A device and method utilizing a computing and control unit, sensors, and nozzles to detect crop plants, calculate potential wetting, and control liquid application to avoid exceeding predefined thresholds, minimizing contact with crops.

Benefits of technology

Minimizes negative effects on crops by precisely applying liquids, reducing phytotoxic reactions and ensuring safe herbicide use, thereby maintaining crop health and yield.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026062470_12112026_PF_FP_ABST
    Figure EP2026062470_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of precision agriculture. The subject matter of the present invention is a device and a method for the targeted application of a liquid to an agricultural field, with the aim of reducing or preventing negative effects on the crop plant.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] BYC250035 Foreign Countries

[0002] DEVICE FOR THE GEREMITED APPLICATION OF A LIQUID ONTO A FOUNDLY PRODUCTIVE FIELD

[0003] The present disclosure relates to the technical field of precision agriculture. The invention relates to a device and a method for the targeted application of a liquid to an agricultural field, aimed at reducing or avoiding negative effects on the crop.

[0004] The term precision agriculture refers to methods of site-differentiated and targeted management of agricultural land.

[0005] One task in precision agriculture is to identify objects such as plants, pests, or companion plants in a digitally captured image, then make a diagnosis, and subsequently carry out an action (e.g., treatment). An example of such a task is "ultra-high-precision spraying" (UHPS), in which a tractor drives across a planted field, taking successive images of the agricultural field with an attached sprayer. Image processing algorithms identify, for example, the companion plants in the images, and then, in a treatment step, spray them with a herbicide using a spray boom. However, in this process, there is also the possibility that the sprayer will hit not only the identified weeds but also neighboring crops.When a herbicide is applied to a crop, phytotoxic reactions can occur, depending on the specific mode of action of the herbicide. These reactions can cause physiological shock, which impairs the normal growth development of the crop. The duration of this growth impairment varies depending on the type of herbicide, the application rate, environmental conditions, the growth stage, and the sensitivity of the crop. In some cases, this shock can last for several days to weeks, potentially leading to yield losses. The risk is particularly high when non-selective herbicides are used, as even small amounts of these can cause significant damage to crops or even kill them.

[0006] There is therefore a need in this technical field to offer solutions for the targeted application of liquids to an agricultural field, where the negative effects on crops are kept to a minimum, so that phytotoxic reactions and the associated physiological shocks in crops can be minimized or ideally avoided. The present invention addresses this problem. A first object of the present invention relates to a device for applying a liquid to an agricultural field comprising

[0007] a computing and control unit,

[0008] at least one storage container to hold the liquid,

[0009] at least one nozzle

[0010] Means for conveying the liquid from the at least one storage container towards the at least one nozzle,

[0011] at least one sensor unit that can be oriented towards the agricultural field and is configured to produce at least one recording,

[0012] wherein the computing and control unit is configured (a) to activate the at least one nozzle to apply the liquid to the agricultural field,

[0013] wherein the computing and control unit is configured, (b) to cause the at least one sensor unit to produce at least one recording and to receive at least one recording, wherein the computing and control unit is configured, (c) to detect at least one crop plant on the recording - if present - and to determine at least approximately the surface area of ​​the crop plant,

[0014] wherein the computing and control unit is configured (d) to calculate with how much liquid an arbitrary area at any distance from the nozzle center of the at least one nozzle will be sprayed in the near future,

[0015] wherein the computing and control unit is configured (e) to determine, at least partially, the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle, based on the information from step (c) and (d), wherein the computing and control unit is configured (f) to determine, at least partially, based on the information from step (e), whether the potential wetting of the surface of the crop plant exceeds a predefined threshold,

[0016] wherein the computing and control unit is configured (g) to direct the at least one nozzle such that the predefined threshold in step (f) is not exceeded. A further object of the present invention relates to a method for applying a liquid to an agricultural field comprising the steps:

[0017] in step (1) moving a device for spreading a liquid on an agricultural field,

[0018] in step (2) conveying the liquid from at least one reservoir towards at least one nozzle while moving the device and activating the at least one nozzle to apply the liquid to the agricultural field, in step (3) causing a sensor unit of the device to generate at least one recording and forwarding this at least one recording to the computing and control unit of the device,

[0019] In step (4) identifying at least one crop plant in the image – if present – ​​and at least approximately determining the surface area of ​​the crop plant; in step (5) calculating how much liquid will be sprayed with any area at any distance from the nozzle center of the at least one nozzle in the near future; in step (6) determining the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle, at least partially based on the information from steps (4) and (5).

[0020] In step (7) determining whether the potential wetting of the surface of the crop exceeds a predefined threshold, at least partially based on the information from step (6),

[0021] In step (8) the at least one nozzle is controlled so that the predefined threshold in step (7) is not exceeded.

[0022] The invention is explained in more detail below, without distinguishing between the invention's subject matter (device, method). Rather, the following explanations are intended to apply analogously to all invention's subject matter, regardless of the context (device, method) in which they are made. For example, a step is described for the device which also applies analogously to the method.

[0023] If the present description or the claims mention steps in a sequence, this does not necessarily mean that the invention is limited to that sequence. Rather, it is conceivable that the steps could also be carried out in a different sequence or even in parallel; unless one step builds upon another, which makes it essential that the building step be carried out subsequently (which will be clear in the specific case). The sequences mentioned thus represent preferred embodiments of the invention.

[0024] The present invention relates to a device for applying a liquid to an agricultural field. The liquid can, for example, be applied selectively to companion plants, thereby reducing or avoiding negative effects caused by the liquid coming into contact with nearby crops. It is also conceivable that the device applies the liquid over the entire area of ​​the agricultural field, while minimizing contact with the crops.

[0025] The liquid can be water, an aqueous solution, or a suspension. The aqueous solution or suspension may contain one or more plant protection products.

[0026] The term "plant protection product" refers to a substance used to protect plants or plant products from harmful organisms or to prevent their effects, to destroy unwanted plants or plant parts, to inhibit or prevent unwanted plant growth, and / or to influence plant life processes in a way other than by providing nutrients (e.g., growth regulators). Examples of plant protection products include herbicides, fungicides, and other pesticides (e.g., insecticides). In one embodiment of the invention, the liquid comprises at least one plant protection product, preferably at least one herbicide.

[0027] Selective herbicides are designed to target specific weed species while largely sparing crops. However, phytotoxic effects can also occur with selective herbicides, particularly if the dosage is not followed as directed and is too high, or if the crop is sensitive. In such cases, physiological shock can occur, impairing crop growth and weakening their health. Therefore, it is important to adhere to the recommended application rate and consider the specific sensitivity of the crop to avoid unwanted damage. In contrast, non-selective herbicides have a broad spectrum of activity and kill all plants they come into contact with, regardless of species. They are often used for complete weed control but can also damage neighboring crops.Therefore, special care must be taken when using non-selective herbicides.

[0028] In one embodiment of the invention, the liquid comprises at least one selective or one non-selective herbicide, preferably at least one selective herbicide.

[0029] The term "agricultural field" refers to a spatially definable area of ​​the earth's surface that is used for agricultural purposes, by planting crops, supplying them with nutrients if necessary, and harvesting them.

[0030] The term "plant" refers to living organisms belonging to the flora category that are capable of photosynthesis. This includes a wide variety of species, including cultivated plants and companion plants, which either occur naturally in a particular ecosystem or are specifically cultivated to meet human needs.

[0031] The term "cultivated plant" refers to a plant that is intentionally grown as a useful or ornamental plant through human intervention.

[0032] The term "companion plants" (often also referred to as weeds) refers to plants of the spontaneous accompanying vegetation (segetal flora) in cultivated crops, grassland, or gardens that are not intentionally cultivated and develop, for example, from the soil's seed pool or through wind dispersal. The term "processing unit" refers to components of a computer or processor used to perform arithmetic and logical operations, as well as to control the execution of instructions and operations. In modern computers, the processing unit and the control unit are often closely integrated and together form the so-called "CPU" ("Central Processing Unit"). These two units work together to enable the processing and execution of instructions in a computer. It is also possible that at least certain functions are performed via cloud-based computing.

[0033] The liquid is stored in a reservoir before application. Multiple reservoirs may be present. Several (different) liquids may be applied. The liquid is applied to the agricultural field via one or more nozzles. In one embodiment, the at least one nozzle is a nozzle with a narrow spray pattern (i.e., a narrow radial circumference of the spray cone) and a small sigma (σ; σ denotes the distribution of droplets across the width of the spray cone). The sigma value depends on the specific geometry and material properties of the at least one nozzle. In addition to the sigma value, the spray pattern also depends on the opening angle of the at least one nozzle and the distance of the nozzle from the agricultural field or the target objects to be sprayed within the agricultural field.In one embodiment, the at least one nozzle has an opening angle of between 10° and 60°, more preferably between 20° and 40°. In another embodiment, the distance from the at least one nozzle to the agricultural field is between 10 cm and 1 m, preferably between 10 cm and 40 cm. Nozzles suitable for use according to the invention are available from various manufacturers such as TeeJet Technologies, Lechler, Hypro, etc. In a further embodiment, several nozzles are used, which are preferably enclosed by a bar or strip (hereinafter referred to as a "spray bar").

[0034] In a further preferred embodiment, the device according to the invention comprises a plurality of nozzles. The term plurality preferably means at least ten, more preferably at least twenty. The nozzles are preferably arranged such that each nozzle applies liquid in an area with a maximum lateral dimension of less than 20 cm, preferably less than 10 cm, most preferably less than 5 cm. The plurality of nozzles can, for example, be arranged side by side along a spray bar that extends transversely (e.g., at an angle of 90°) to the direction of movement of the device. The distance between two adjacent nozzles can, for example, be between 1 cm and 10 cm, and more preferably between 2 and 5 cm. Each nozzle can be assigned at least one sensor unit (e.g., a camera). For application, the liquid is conveyed by means of conveying means from the at least one reservoir towards the at least one nozzle.A pump can be used, for example, to move the liquid.

[0035] The device according to the invention comprises at least one sensor unit. A sensor unit comprises at least one sensor. A "sensor" is a technical component that can qualitatively or quantitatively detect certain physical and / or chemical properties and / or the material composition of its environment. These quantities are detected by means of physical or chemical effects and converted into a further processable, usually electrical or optical, signal. A sensor unit may contain means for processing signals supplied by the at least one sensor. A sensor unit may include means for transmitting and / or forwarding signals and / or information (e.g., to the processing and control unit). The one or more sensor units may be part of the system and / or be connected to it via a communication link (e.g., via radio or cable).The one or more sensor units can be configured to continuously, at defined intervals, or upon the occurrence of defined events, transmit one or more signals to the processing and control unit, based on which the processing and control unit performs further steps. The at least one sensor unit (hereinafter also referred to as the "recording sensor unit"), which can be oriented towards the agricultural field, preferably comprises one or more cameras. That is, the term "recording" preferably describes "image capture." The camera axis of the at least one camera as part of the device according to the invention can preferably be oriented substantially perpendicular (vertically) to the ground (i.e., the at least one camera points "downwards" or "straight ahead" with respect to the ground). Preferably, the at least one camera is located (in the direction of travel) in front of the at least one spraying unit.This distance is preferably between 30 cm and 100 cm, and even more preferably between 40 and 60 cm. The camera used according to the invention can comprise an image sensor and optical elements. The image sensor is a device for capturing two-dimensional images of light electrically. Typically, these are semiconductor-based image sensors such as CCD (CCD = charge-coupled device) or CMOS (CMOS = complementary metal-oxide semiconductor) sensors. The optical elements (lenses, apertures, and the like) serve to produce the sharpest possible image of an object on the image sensor.

[0036] The camera can be instructed by the processing and control unit to generate images of the agricultural field at defined intervals or continuously. The processing and control unit can be the processing and control unit of the device according to the invention or a separate processing and control unit. Preferably, the processing and control unit of the device according to the invention is used for this purpose. Each image can be marked with a timestamp to document the exact time the image was taken. It is also possible that no crop (or parts thereof) is present in a single image and, for example, only the soil or the agricultural field, or only a companion plant, is captured.

[0037] The device according to the invention can, in addition to the at least one sensor unit described above, also include at least one positioning unit. The at least one positioning unit can also include one or more sensors. The at least one positioning unit of the device is configured such that, at the time the at least one sensor unit takes a picture, the position data of the at least one sensor unit (i.e., the camera) are acquired and this information can be transmitted to the processing and control unit of the device. The at least one positioning unit can, for example, include a receiver of a satellite navigation system (GNSS, Global Navigation Satellite System), also commonly referred to as a GPS receiver. The Global Positioning System (abbreviated GPS)GPS (officially NAVSTAR GPS) is an example of a global satellite navigation system for determining position; other examples include GLONASS, Galileo, and BeiDou. The satellites of such a satellite navigation system communicate their precise position and time via radio codes. To determine a position, a receiver (the "GPS receiver") must receive signals from at least four satellites simultaneously. The receiver measures the pseudo-signal travel times and uses these to calculate the current position.

[0038] A time and location stamp can now be captured by the positioning unit, determining the positional coordinates of an object in real time. This data can be georeferenced with the images and the crop plants (or companion plants) identified within them by synchronizing them with the timestamps of the images. This allows a precise geographic position to be assigned to the crop plant (or companion plant) identified in the image.

[0039] In one embodiment, the computing and control unit is configured (in step a) to activate the at least one nozzle to apply the liquid to the agricultural field. Applying a liquid to an agricultural field is accomplished using at least one nozzle that enables spraying. The nozzle breaks the liquid down into fine droplets and distributes them evenly, thus ensuring targeted application. Several scenarios are conceivable when applying a liquid containing a herbicide to an agricultural field. One possibility is the complete application of the liquid, in which the entire field—except for the cultivated plants—is treated evenly to control weeds. According to the invention, the aim in this scenario is to avoid spraying the cultivated plants as much as possible to prevent any negative effects on their growth.Another method is targeted spraying of companion plants, where only the identified companion plant populations are treated. In this scenario as well, the cultivated plants should ideally not be sprayed, so that negative effects on these plants can be minimized or prevented.

[0040] The computing and control unit of the device is configured (in step b) to cause the at least one sensor unit to generate at least one image and to receive this at least one image of the agricultural field. In a preferred embodiment, the computing and control unit is configured in step (b) to continuously generate image images of the agricultural field. The at least one sensor unit is preferably located on or attached to an agricultural device that is moving or movable within the agricultural field. In a further preferred embodiment, an image generated in step (b) represents an area that is preferably located in the direction of movement of the device in front of the at least one nozzle.

[0041] The processing and control unit is configured (in step c) to process the received image, preferably using image processing, in order to recognize at least one cultivated plant—if present—in the image, e.g., a defined cultivated plant or at least part of a cultivated plant. Methods and devices for recognizing objects in image images are described in detail in the prior art (see, e.g., W02020120802A1, W02020120804A1, WO2020229585A1). Various methods can be used for image processing for object recognition, such as feature extraction with deep learning models like neural networks, support vector machines (SVMs), decision trees, etc.Other possible methods that can be used include color-based recognition, texture analysis, shape-based recognition, spectral analysis, color histogram analysis, the histogram of oriented gradient (HOG) method, morphological operations, or a combination of these methods.

[0042] In one embodiment, the computing and control unit is configured to perform machine learning to detect at least one crop plant—if present—in the image. The models used in this image processing (hereinafter also referred to as "image processing methods") are preferably based on deep learning techniques, such as neural networks (for example, a convolutional neural network (CNN)). Such a model is trained with annotated images to learn how to distinguish between different classes, such as a crop plant and a companion plant. After training, such a model can be applied to new images. An image received by the at least one sensor unit can be preprocessed for use with the model.Preprocessing includes common methods such as normalization or scaling steps, noise removal, contrast adjustments, and other methods to improve data quality, as well as converting the image to the model's input format. The model assigns each pixel in the image to a specific class, for example, whether the pixel belongs to a crop plant, a companion plant, or the background. In this step, different plant parts (leaf, stem, flower) of the crop plant can also be identified. This process is also known as semantic segmentation. Semantic segmentation for identifying plants in a field is described in the prior art (see, for example, WO19226869A1). Postprocessing of the classified data at the pixel level of the model can include transformations or analysis operations such as generating pixel masks (e.g.,The following steps can be performed: binary representation of pixels belonging to or not belonging to the recognized crop plant, generation of bounding boxes (rectangular frames around a recognized crop plant), determination of the centroid (the center of the recognized crop plant's area, which can be used as a representative point for the recognized crop plant), and generation of an outline polygon (a series of points describing the contour of the recognized crop plant). In addition to semantic segmentation, object detection algorithms such as YOLO, SSD, Faster-RCNN, etc., can also be used. With these algorithms, it is possible to directly generate bounding boxes or outline polygons for the recognized crop plant, instead of using pixel masks.

[0043] Preferably, the image processing methods detect at least one entire crop plant. However, it can happen that no crop plant or only a part of one is present in the image, and only this part can be detected. The image processing methods mentioned above, including the pre- and post-processing steps, can also be performed if only part of a crop plant is detected in the image. In addition to detecting the at least one crop plant in the at least one image, the surface of the detected crop plant is also determined, at least approximately, in step c. In a preferred variant, the leaf surface of the crop plant is determined as the surface of the at least one crop plant. When the camera is moved along an agricultural device, the leaf surfaces are primarily captured from an overhead perspective, with the stem of the crop plant generally not in the field of view.This arrangement enables precise analysis of the leaves relevant for the application of pesticides. In another embodiment, flowers on a crop can be detected. If crops in the flowering stage are identified, this can lead to the decision to abort the spraying process.

[0044] The area of ​​a bounding box can be easily calculated, for example, by multiplying its width and height. However, this method only provides an approximation of the surface area (e.g., an upper limit for the surface area) of the crop, as it does not take the actual shape of the crop into account. It is important to note that the aperture angle of the camera lens, the camera resolution, and the camera height all affect how many pixels correspond to one millimeter. These parameters determine the spatial resolution and thus the accuracy of the image measurements. When using an outline polygon, the surface area of ​​the crop can be calculated using algorithms such as the Shoelace algorithm or the Monte Carlo method, either by counting the pixels within the outline polygon or by directly calculating the area of ​​the polygon.This method provides a more accurate estimate of the actual surface area because it takes the shape of the crop plant into account. Starting with an outline polygon, the 2D surface area of ​​the crop plant can generally be calculated. The outline polygon represents a two-dimensional representation of the contours and allows for the calculation of the area within these contours. The centroid of the identified crop plant, as described above, can also be used for area calculation, especially in combination with other methods. While the centroid itself does not calculate an area, it can be useful for analyzing the distribution of the area and gaining information about the shape and symmetry of the identified crop plant.

[0045] In a further embodiment of the invention, step (c) also determines the distance between the at least one detected crop plant and the recording sensor unit (or the at least one nozzle that could potentially wet the detected crop plant). This can be done using the previously described georeferenced data of the image on which the crop plant was detected, as well as knowledge of the optical parameters of the recording sensor unit (e.g., aperture angle, vertical and horizontal resolution) and the distance of the recording sensor unit to the soil of the agricultural field, and knowledge of the substantially constant distance between the recording sensor unit and the at least one nozzle (see also the details in PCT / EP2025 / 050242). In one embodiment of the invention, the at least approximate determination of the surface area of ​​the crop plant refers to the leaf surface of the crop plant.In another embodiment, the at least approximate determination of the surface area of ​​the cultivated plant refers to a 2D surface area calculation of at least one cultivated plant. This 2D surface area calculation is preferably performed starting from a bounding box and even more preferably from an outline polygon.

[0046] In a further step, the computing and control unit is configured (in step d) to calculate how much liquid will be sprayed onto an arbitrary area at any distance from the nozzle center of the at least one nozzle in the near future. In a preferred embodiment, at least one model is used to determine the intensity distribution of the at least one nozzle. Such a model is briefly explained below by way of example, noting that there are different equations for different nozzle types. A person skilled in the art can adapt the calculation to other nozzle types based on the equations discussed below by way of example. The spray pattern distribution of the at least one nozzle resembles a Gaussian bell curve, which has a characteristic bell shape. In this distribution, the maximum intensity lies directly under the spray nozzle, where the concentration of spray particles is highest.As the distance from the center increases, the intensity of the spray particles decreases, resulting in a downward-sloping curve that describes the spray intensity distribution. Therefore, for calculating the spray intensity of any area outside the region of the Gaussian bell curve, it can be assumed that no liquid from the at least one nozzle strikes this area. A possible mathematical model for describing the spray intensity 1(d) as a function of distance (d) is formula (1):.

[0047]

[0048] (1)

[0049] where Ipeak is the maximum intensity at the nozzle center and o (sigma) represents the standard deviation of the distribution, which describes the width of the bell curve. The nozzle center is defined as the perpendicular line running in the spray direction of the nozzle and represents the optimal point for the application of the spray mist. This line represents the area where the liquid with the highest concentration and uniform distribution hits the target surface. The determination of I pea k can be calculated using the equations according to formulas (2) to (5) as follows:

[0050] Q = AR ■ i> ■ W

[0051] (2)

[0052] Where Q is the flow rate, AR the application rate (e.g., in 1 / ha), v the velocity of the spray particles, and W the width of the spray jet. W can be obtained using the trigonometric equation of formula (3): W = 2 x height x tan (0 / 2), where height describes the height of the nozzle to the crop and 0 the spray angle of the nozzle.

[0053] The flow rate Q can be calculated as the integral of the Gaussian intensity distribution over the entire spray range. For this purpose, the intensity distribution 1(d), described by the Gaussian function, is transformed into an integration form to capture the total amount of spray material distributed over a given area. Formula (4) for the Gaussian intensity is:

[0054]

[0055] (4)By equating the two formulas (2) and (4) for Q, formula (5) is obtained:

[0056] ft r 11 'A tT

[0057]

[0058] (5)

[0059] This means that I pea k can be calculated from the equation in formula (5) mentioned above, where W, AR, and o have the meanings described above (see formulas 1 to 3). In this context, it should be noted that nozzle characteristics can also be used to improve the calculation. For example, the intensity distribution can change with increasing pressure (in the spray lines), leading to a pressure dependence of sigma. The formulas mentioned above still apply, but the parameter sigma can change.

[0060] Furthermore, it should be noted that the pressure (in the spray lines) and the speed of the spraying device must be selected so that the desired application rate can be achieved; that is, the nozzle (at least one), the pressure, and the driving speed of the spraying device must be properly coordinated. Modern spraying devices adjust the pressure depending on the speed to maintain a constant application rate. This adjustment has become standard in modern spraying devices to ensure a uniform distribution of pesticides or fertilizers and to avoid over- or under-application.

[0061] In one example, the term "in the near future" in step (d) encompasses a period until steps (b) and (c) have been performed for a further recording (i.e., the subsequent recording). This would be, for example, a period of 10 to 500 milliseconds.

[0062] Another aspect of the invention relates to the device, wherein in step (e) the computing and control unit is configured to determine, at least partially, the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle, based on the information from steps (c) and (d).

[0063] In one embodiment, the amount of liquid that the surface of a detected crop plant receives at a distance d is calculated as follows: Liquid wetting on the surface of a detected crop plant V = 1(d) x A (Formula 6), where 1(d) can be determined as specified in formula (1) above, and A is the surface area of ​​the detected crop plant, determined at least approximately (see step c). Here, d is the distance from the nozzle center to the leaf edge of the crop plant. This estimate for V is an upper limit, since the intensity 1(d) decreases as d increases.

[0064] In a further embodiment of the invention, the amount of liquid F that the surface of a detected crop plant receives at a distance d can be calculated by determining the function 1(d) over the exact shape and distance (or an approximate shape such as a circle) at a specific distance using a numerical integration method. This method is generally more accurate but also computationally more expensive, which is a critical factor in a real-time application.

[0065] Applying a specific amount of liquid (preferably a plant protection product) to the surface of a crop using a sprayer does not necessarily mean that the entire amount will penetrate the plant. A portion of the spray is typically lost, a phenomenon known as "drift" and "beading off." These losses can be influenced by various factors, such as droplet size, plant surface texture, and environmental conditions. However, it is possible to anticipate these losses, as relevant data and experience exist regarding application methods and the specific properties of the plant protection product used.Therefore, in one embodiment of the invention, the term “potential wetting of the surface of the crop” can be approximated as the cumulative amount of liquid that could come into contact with the surface of a detected crop. To determine the cumulative amount that could come into contact with the surface of the crop during the spraying process, the formula (6) described above: V = 1(d) x A can be used.

[0066] Another aspect of the invention relates to the device, wherein in step (f) the computing and control unit is configured to determine, at least partially, on the basis of the information from step (e), whether the potential wetting of the surface of the crop exceeds a predefined threshold.

[0067] Determining a predefined threshold, above which the applied quantity of a liquid—preferably comprising a plant protection product—becomes phytotoxic to a crop and impairs its growth, can be achieved in one or more steps. In one example, the predefined threshold is determined by a threshold analysis of the at least one plant protection product present in the liquid. This threshold analysis includes experimental methods and / or mathematical modeling. In this step, existing data on the phytotoxic effects of the plant protection product (e.g., from registration studies and the product label) on the specific crop are evaluated. This data can then be integrated into mathematical models to quantitatively describe the relationship between the applied quantity and the observed phytotoxic effects.Mathematical methods that establish a relationship between the amount of pesticide applied and its phytotoxic effects on crops include regression analysis, which is frequently used to quantify the dose-response relationship. Dose-response models, such as the logistic model, describe the relationship between the amount of pesticide and the corresponding phytotoxic effects. Statistical hypothesis tests, such as the t-test or ANOVA, can be used to examine significant differences in effects at different application rates. Furthermore, calculating correlations allows for the determination of the strength and direction of the relationship between the variables. Mathematical simulations, often based on differential equations, can also investigate the dynamics of chemical interactions.Based on the results of the mathematical models, a (predefined) threshold value can be established. This threshold value can include a safety margin to account for unforeseen variability. According to the invention, several (predefined) threshold values ​​can also be used. These threshold values ​​can be determined, for example, for different pesticides or combinations thereof and / or different crops.

[0068] Optionally, in addition to the existing data described above, toxicological studies can be conducted in which different concentrations of the liquid containing the plant protection product are applied to the crop to observe the effects on growth and health, documenting typical phytotoxic symptoms such as leaf discoloration or growth inhibition. The data obtained help to determine the No Observed Adverse Effect Level (NOAEL) or the Lowest Observed Adverse Effect Level (LOAEL). Furthermore, field trials can also be carried out to test the effects under realistic conditions and to take environmental factors into account. These data can then be integrated into the mathematical models mentioned above, in addition to the known data, to quantitatively describe the relationship between the applied quantity and the observed phytotoxic effects even more precisely.

[0069] A further aspect of the invention relates to the device, wherein in step (g) the computing and control unit is configured to control at least one nozzle such that the predefined threshold in step (f) is not exceeded. The term "control" preferably encompasses various scenarios, such as the inactivation of the at least one nozzle, which in turn preferably includes deactivation or non-activation of the at least one nozzle. In other words, the at least one nozzle can be active and deactivated by the calculation in step (f) and the control of the nozzle in step (g) (if the predefined threshold is exceeded). In another scenario, the at least one nozzle can be inactive and remain inactive by the calculation in step (f) and the control of the nozzle in step (g) (i.e., not be activated if the predefined threshold is exceeded).In another scenario, at least one nozzle can remain active or be activated in step (g) if the predefined threshold is not exceeded.

[0070] A further aspect of the invention relates to the device, wherein the computing and control unit is configured in step (c) to detect at least one companion plant on the recording and to activate the at least one nozzle to control the detected companion plant with the liquid, provided that the predefined threshold for at least one crop plant located near the companion plant and detected in step (f) is not exceeded. The detection of the at least one companion plant on the at least one recording is carried out analogously to the above described procedure for the at least one crop plant. The control of a detected companion plant in UHPS is known to those skilled in the art and described in the prior art (see, for example, PCT / EP2025 / 050242). The term "crop plant located near the companion plant" preferably refers to plants that are entirely or partially contained within the same recording.have been at least partially recognized.

[0071] The device according to the invention can be part of a spraying device, for example, an agricultural machine, a robot, or an aircraft (e.g., a drone), or can be connected to one. Such a spraying device can move autonomously in or above a field or be controlled by a person. Preferably, the device is part of a ground-level spraying device (i.e., the at least one nozzle is located at or less than 1 meter above the agricultural field) or can be connected to one.

[0072] In another embodiment, the described steps (a) to (g) or (2) to (8) are carried out iteratively.

[0073] The steps described above can also be performed when using multiple nozzles. As already described, the use of nozzles with a narrow spray pattern is preferred. With these nozzles, the overlap of the spray patterns from multiple nozzles on a spray bar is negligible. In an exemplary embodiment, the distance of the detected crop to the nearest open nozzle located to the left of the crop on the spray bar in the direction of travel of the device, as well as to the next nearest open nozzle located to the right of the crop on the spray bar, can be determined while passing over the crop. The left and right nozzles are arranged laterally adjacent to each other on the spray bar. The quantities obtained from both nozzles can then be added, for example, to calculate the upper limit for the total application.This overall application can be used in step (f) to determine whether the potential wetting of the crop surface exceeds a predefined threshold. If so, at least one (or both) of these nozzles can be controlled in step (g) so that the predefined threshold is not exceeded. For example, deactivating just one nozzle might be sufficient to keep the threshold below the threshold. However, deactivating both nozzles may be necessary.

[0074] If, however, the detected crop is sprayed by a larger number of nozzles (which would be undesirable), the intersection area of ​​the spray patterns could be approximated, and the amount of liquid that reaches the surface of the detected crop could be estimated, taking into account the liquid application rate. This value could then be used analogously to step (f) above to check whether the predefined threshold is exceeded. Subsequently, depending on requirements, one or more nozzles can be controlled so that the predefined threshold is not exceeded (in step g). Embodiments of the present invention are:

[0075] Figures 1 to 5 show exemplary and schematic embodiments of the device and method according to the invention.

[0076] Figure 1 shows the device 10 for applying a liquid F to an agricultural field LF. The device 10 comprises a computing and control unit 11, at least one reservoir 12 for receiving the liquid F, at least one nozzle 13, means 15 for conveying the liquid F from the at least one reservoir 12 towards the at least one nozzle 13, and at least one collection unit 16, which can be oriented towards the agricultural field F. In Figure 1, the device 10 moves from right to left, and the collection unit 16 is located in the direction of movement in front of the at least one nozzle 13. The direction of the nozzle 13 is shown obliquely relative to the ground surface. However, the direction of the nozzle 13 can also be directed straight downwards relative to the ground surface.

[0077] Figure 2 shows an example of the device 10 for applying a liquid to an agricultural field. The device 10 comprises a computing and control unit, at least one reservoir for receiving the liquid, at least one nozzle, means for conveying the liquid from the at least one reservoir towards the at least one nozzle, at least one sensor unit that can be oriented towards the agricultural field, and is configured to generate at least one image. In step (a), the computing and control unit is configured to activate the at least one nozzle to apply the liquid to the agricultural field. In step (b), the computing and control unit is configured to cause the at least one sensor unit to generate at least one image and to receive this at least one image.In step (c), the processing and control unit is configured to detect at least one crop plant in the image—if present—and to determine, at least approximately, the surface area of ​​the crop plant. In step (d), the processing and control unit is configured to calculate how much liquid will be sprayed with any area at any distance from the nozzle center of the at least one nozzle in the near future. In step (e), the processing and control unit is configured, at least partially, to determine the potential wetting of the crop plant's surface with the liquid from the at least one nozzle, based on the information from steps (c) and (d). In step (f), the processing and control unit is configured, at least partially, to determine, based on the information from step (e), whether the potential wetting of the crop plant's surface exceeds a predefined threshold.In step (g), the computing and control unit is configured to control at least one nozzle in such a way that the predefined threshold in step (f) is not exceeded. The dashed arrows in the upper right indicate that steps a) to g) are preferably performed iteratively.

[0078] Figure 3 shows an example of a method 100 for applying a liquid to an agricultural field. The method 100 comprises the following steps (1) to (8). In step (1), a device for applying a liquid to an agricultural field is moved. In step (2), the liquid is conveyed from at least one reservoir towards at least one nozzle while moving the device, and the at least one nozzle is activated to apply the liquid to the agricultural field. In step (3), a sensor unit of the device is caused to generate at least one image and to transmit this image to the device's computing and control unit. In step (4), at least one crop is detected in the image, and the surface area of ​​the crop is determined, at least approximately.Step (5) calculates how much liquid will be sprayed with an arbitrary area at any distance from the nozzle center of the at least one nozzle in the near future. Step (6) determines the potential wetting of the crop surface with the liquid from the at least one nozzle, at least partially, based on the information from steps (4) and (5). Step (7) determines whether the potential wetting of the crop surface exceeds a predefined threshold, at least partially, based on the information from step (6). Step (8) controls the at least one nozzle so that the predefined threshold in step (7) is not exceeded. The dashed arrows in the upper right indicate that steps 2) to 8) are preferably performed iteratively.

[0079] Figure 4 shows an example of the spray distribution of nozzles with different sigma (o) values. The ordinate (y-axis) represents the spray intensity (7) in pL / m². 2The graph shows the distance to the nozzle center (d) in centimeters (cm) on the abscissa (x-axis), with zero representing the nozzle center. For simplicity, this graph assumes that the maximum intensity (Ipeak) at the nozzle center is the same for all nozzles. The curve with the black circle represents the spray pattern from a commercially available nozzle with a sigma value (o) of 1.45 cm (o = spray distance W / 6). The curve with the black square represents the spray pattern from a commercially available nozzle with a sigma value (o) of 2.90 cm (o = 2 x (W / 6)). The curve with the black triangle represents the spray pattern from a commercially available nozzle with a sigma value (o) of 4.35 cm (o = 3 x (W / 6)).For a nozzle positioned 50 cm from the agricultural field (height) and with a nozzle angle (θ) of 10°, the spray range (W) can be calculated using the formula: 2 x height x tan (θ / 2), which approximately equals 8.7 cm. With a sigma value (θ) of 1.45 cm, 99.7% of the sprayed liquid is contained within the spray range (W) of 8.7 cm. With a sigma value (θ) of 2.90 cm, 95% of the sprayed liquid is contained within the spray range (W) of 8.7 cm. With a sigma value (θ) of 4.35 cm, 68% of the sprayed liquid is contained within the spray range (W) of 8.7 cm.

[0080] Figure 5 shows an example of the spray distribution from the nozzles already described in Figure 4. The spray intensity (7) in pL / m² is shown on the ordinate (y-axis). 2The abscissa (x-axis) shows the distance to the nozzle center (d) in meters (m), with zero representing the nozzle center. This representation shows the actual maximum intensities (Ipeaks) at the nozzle center of all three nozzles already described in Figure 4. pea ks can be determined, for example, by using the formulas (2) to (5) described above and the same data as described for Figure 4. For the curve with the black circle, a maximum intensity I is obtained. pea k) at the nozzle center of 4.79 pL / m 2 (Nozzle with a sigma value (θ) of 1.45 cm). For the curve with the black-filled square, a maximum intensity I is obtained. pea k) at the nozzle center of 2.39 pL / m 2 (Nozzle with a sigma value (θ) of 2.90 cm). For the curve with the black-filled triangle, a maximum intensity (I) is obtained. pea k) at the nozzle center of 1.60 pL / m2 (Nozzle with a sigma value (o) of 4.35 cm). The use of a nozzle with a sigma value (o) of 1.45 cm (black filled circle) is preferred according to the invention as described in this application.

Claims

PATENT CLAIMS 1. Device for applying a liquid to an agricultural field comprising a calculating and control unit, at least one storage container to hold the liquid, at least one nozzle Means for conveying the liquid from the at least one storage container towards the at least one nozzle, at least one sensor unit that can be oriented towards the agricultural field and is configured to produce at least one recording, wherein the computing and control unit is configured (a) to activate the at least one nozzle to apply the liquid to the agricultural field, wherein the computing and control unit is configured, (b) to cause the at least one sensor unit to produce at least one recording and to receive at least one recording, wherein the computing and control unit is configured to (c) detect at least one crop plant in the recording - if present - and to determine at least approximately the surface area of ​​the crop plant, wherein the computing and control unit is configured (d) to calculate with how much liquid an arbitrary area at any distance from the nozzle center of the at least one nozzle will be sprayed in the near future, wherein the computing and control unit is configured (e) to determine, at least partially, the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle, based on the information from step (c) and (d), wherein the computing and control unit is configured (f) to determine, at least partially, based on the information from step (e), whether the potential wetting of the surface of the crop plant exceeds a predefined threshold, wherein the computing and control unit is configured (g) to direct the at least one nozzle such that the predefined threshold in step (f) is not exceeded.

2. Device according to claim 1, wherein in step (b) continuous image recordings of the agricultural field are generated by an agricultural device moving thereon.

3. Device according to claim 2, wherein in step (b) an image capture represents an area which is preferably located in the direction of movement of the device in front of the at least one nozzle.

4. Device according to any one of claims 1 to 3, wherein in step (c) the recognition of the at least one crop plant on the image and the at least approximate determination of the surface of the at least one crop plant are carried out by using at least one image processing method.

5. Device according to any one of claims 1 to 4, wherein in step (c) the at least approximate determination of the surface of the at least one crop plant includes at least the leaf surface of the crop plant.

6. Device according to any one of claims 1 to 5, wherein in step (d) the term "near future" includes a period of time until steps (b) and (c) have been carried out for a further recording.

7. Device according to any one of claims 1 to 6, wherein in step (d) at least one model is used to determine the intensity distribution of the at least one nozzle.

8. Device according to any one of claims 1 to 7, wherein in step (e) determining the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle comprises determining the cumulative amount of liquid that could come into contact with the surface of the at least one crop plant.

9. Device according to any one of claims 1 to 8, wherein the liquid comprises at least one plant protection product, preferably a herbicide.

10. Device according to claim 9, wherein step (f) the predefined threshold is determined by a threshold analysis of the at least one pesticide present in the liquid and the threshold analysis comprises experimental methods and / or mathematical modeling.

11. Device according to one of claims 1 to 10, wherein in step (g) the control of the at least one nozzle comprises an inactivation of the at least one nozzle, wherein the inactivation of the at least one nozzle preferably comprises the deactivation or the non-activation of the at least one nozzle.

12. Device according to any one of claims 1 to 11, wherein the computing and control unit is configured in step (c) to detect at least one companion plant on the recording and to activate the at least one nozzle to control the detected companion plant with the liquid, provided that the predefined threshold for at least one crop plant located near the companion plant and detected in step (f) is not exceeded.

13. Device according to any one of claims 1 to 12, wherein the device is an agricultural device and preferably a part of an agricultural machine, a robot or a drone, or is connectable to one.

14. Device according to any one of claims 1 to 13, wherein steps (a) to (g) are performed iteratively.

15. Method for applying a liquid to an agricultural field comprising the steps: in step (1) moving a device for spreading a liquid on an agricultural field, In step (2) conveying the liquid from at least one reservoir towards at least one nozzle while moving the device and activating the at least one nozzle to apply the liquid to the agricultural field, in step (3) causing a sensor unit of the device to generate at least one recording and forwarding this at least one recording to the computing and control unit of the device, In step (4) identifying at least one crop plant in the image – if present – ​​and at least approximately determining the surface area of ​​the crop plant; in step (5) calculating how much liquid an arbitrary area at any distance from the nozzle center of the at least one nozzle will be sprayed with in the near future; in step (6) determining the potential wetting of the surface of the crop plant with the liquid from the at least one nozzle at least partially based on the information from steps (4) and (5). in step (7) Determine whether the potential wetting of the surface of the crop plant exceeds a predefined threshold, at least partially based on the information from step (6), In step (8) the at least one nozzle is controlled so that the predefined threshold in step (7) is not exceeded.