System and method for distributing agricultural material, and agricultural machinery
The system addresses the challenge of monitoring and classifying agricultural material distribution by using a height-adjustable boom and AI-driven evaluation to ensure accurate and efficient material distribution.
Patent Information
- Application Number
- DE102024128776
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-10-07
- Publication Date
- 2026-04-09
AI Technical Summary
Existing agricultural material distribution systems lack effective monitoring and classification of distribution accuracy, particularly in concealed areas, making it difficult to determine if the distribution is correct.
A system with a height-adjustable distribution boom and detection device to record and evaluate the distribution profile of agricultural material, using an evaluation unit to compare it with a reference profile, and an artificial intelligence unit to determine deviations and issue warnings or automatic responses.
Enables accurate monitoring and classification of distribution patterns, ensuring correct material distribution and providing real-time feedback for adjustments, thereby improving distribution efficiency and accuracy.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a system for distributing agricultural material on agricultural land, wherein the system comprises at least one height-adjustable distribution boom with at least two distribution sections movable relative to each other and with a plurality of distribution units arranged on the distribution sections for distributing the material, wherein the material exits each distribution unit with a distribution profile. The invention further relates to a method for distributing agricultural material on agricultural land. In addition, the invention relates to an agricultural machine.
[0002] Systems for distributing materials are known from the prior art. Often, the distribution of the material is carried out purely visually by a person. However, monitoring the distribution of the material is particularly difficult with such systems, especially in concealed areas that a person cannot see.
[0003] The object of the invention is therefore to eliminate the described disadvantages of the prior art. In particular, a system is to be provided by means of which the distribution of the distribution units can be classified as correct or incorrect.
[0004] These problems are solved by a system with the features of independent claim 1, by a method for spreading agricultural material with the features of method claim 16, and by an agricultural machine with the features of claim 17. The invention further relates to a method for generating training data according to claim 19 and to the use of a trained artificial intelligence unit according to claim 20. Advantageous embodiments and further developments of the invention are disclosed in the claims and the following description with at least partial reference to the figures.
[0005] According to one aspect of the invention, a system for distributing agricultural material on an agricultural area is provided, wherein the system comprises at least a height-adjustable distribution boom with at least two distribution sections movable relative to each other and with a plurality of distribution units arranged on the distribution sections for distributing the material, wherein the material exits from each distribution unit with a distribution profile.
[0006] The totality of the distribution profiles, especially at a specific time, can also be referred to as a distribution pattern of the distribution linkage.
[0007] The invention is characterized by - a recording device which is set up to record at least part of the distribution profile of at least one distribution unit, - an evaluation unit which is set up to perform a comparison between the recorded distribution profile and a reference distribution profile.
[0008] Preferably, the system includes the data acquisition device and the evaluation unit.
[0009] Preferably, the agricultural material being distributed is any material that can be spread on agricultural land. Preferably, the agricultural material being distributed is a spray liquid, a fertilizer (preferably a solid, liquid and / or liquefied fertilizer), seeds, and / or the like.
[0010] In principle, a distribution boom is a structure to which distribution units for distributing the material being applied are attached or to which distribution units for distributing the material are coupled. The distribution boom can also be referred to as a spray boom, particularly when applying pesticides. The distribution boom may be height-adjustable. Preferably, the system comprises a support frame to which the distribution boom is mounted, and the support frame is preferably configured to allow the distribution boom to be moved vertically. More preferably, the distribution boom can be rotatably mounted about at least one axis of rotation. Preferably, the distribution boom can also be laterally displaceable. Due to its height adjustability and, if applicable, its rotatability and / or displacement capability, the distribution boom can be well adapted to the geographical requirements of the agricultural area.
[0011] According to the invention, the distribution linkage comprises at least two distribution sections that are movable relative to each other. Preferably, the two distribution sections that are movable relative to each other are designed to be foldable, in particular foldable together. For this purpose, at least one hinge is provided, which is configured so that the two distribution sections can be moved relative to each other. Preferably, the distribution linkage comprises three or more distribution sections, wherein preferably one distribution section is provided as a central distribution section, and the further distribution sections are coupled directly or indirectly, for example by means of further distribution sections, to the central distribution section. Preferably, the central distribution section may be coupled to the support frame.
[0012] The distribution units can be designed according to the material being distributed. For example, and this list is not exhaustive, distribution units on a field sprayer can take the form of distribution nozzles or spray nozzles.
[0013] The distributed material exits the respective distribution unit with a distribution profile. This means that the distribution profile is assigned according to the respective distribution unit. Preferably, the distribution profile is dependent on certain parameters and / or adjustable and / or predefined.
[0014] According to a preferred embodiment, at least one characteristic specifying the distribution profile can be assigned to each distribution profile. A characteristic can be a value that specifies and / or describes the distribution profile. A characteristic specifying the distribution profile can also be referred to as a characteristic or as the characteristic of the distribution profile. The distribution profile can be described in more detail using the at least one characteristic.
[0015] Preferably, at least one characteristic of the distribution profile comprises at least one of the following: opening angle, shape, uniformity of distribution, size, distribution pattern on an agricultural area, material spectrum, in particular droplet spectrum, average particle size of the material, exit velocity of the material from the distribution unit, quantity of material, and / or any combination thereof. In principle, further and / or other characteristics are also conceivable. The foregoing list is not exhaustive. Preferably, the distribution profile can have a spatial distribution, which can be described, for example, as a distribution fan, distribution cone, spray pattern, and / or the like. The spatial distribution can be described by one or more characteristics.
[0016] The system according to the invention is characterized by a detection device configured to at least partially detect the distribution profile of at least one distribution unit. Preferably, the detection device is configured to detect the distribution profile of at least one distribution unit with at least one characteristic that specifies the distribution profile. More preferably, the detection device is configured to detect at least one characteristic of the distribution profile, and more preferably, several or all characteristics of the distribution profile. Preferably, the system comprises the detection device.
[0017] The distribution profile can preferably depend on further parameters, such as those of the system, for example, the pressure applied to each distribution unit, which is used to distribute the material through the distribution units, and / or the geometry and / or design of the distribution unit itself, and / or the type and / or composition of the material being distributed, such as active ingredient, concentration, viscosity, and / or the like. This list is exemplary and not exhaustive; that is, further and / or other parameters are conceivable.
[0018] Preferably, the detection device can be coupled to the distribution linkage. Alternatively, it is conceivable that the detection device is coupled to the support frame. The detection device is positioned close to the distribution units and can effectively detect at least one distribution profile. Furthermore, it is conceivable that the detection device can be positioned at any point within the system and / or on an agricultural machine, or coupled to the system.
[0019] Preferably, the detection device may be arranged in front of and / or behind the distribution unit along a direction of travel. Preferably, the detection device may be arranged above and / or below the distribution unit along a vertical direction. Depending on requirements, space constraints, and / or similar factors, an optimal position for the detection device may be determined.
[0020] Furthermore, the system according to the invention is characterized by an evaluation unit which is configured to determine and compare the recorded distribution profile with a reference distribution profile. Preferably, the system includes the evaluation unit.
[0021] According to a preferred embodiment, the evaluation unit may be configured to compare the acquired distribution profile with a reference distribution profile in order to determine, based on this comparison, whether the acquired distribution profile corresponds to the reference distribution profile. If, based on the comparison, the acquired distribution profile and the reference distribution profile differ, it can be assumed that there is a specific cause that leads to a change or deviation of the acquired distribution profile from the reference distribution profile. If the acquired distribution profile and the reference distribution profile are substantially the same, it can be assumed that the acquired distribution profile is functioning correctly. "Substantially" means that deviations between the acquired distribution profile and the reference distribution profile are tolerable to a certain extent.
[0022] According to a preferred embodiment, the evaluation unit may be configured to determine differences between the recorded distribution profile and the reference distribution profile by comparing them. If the comparison between the recorded distribution profile and the reference distribution profile using the evaluation unit reveals that the recorded distribution profile does not correspond to the reference distribution profile, the evaluation unit can further determine the differences between the recorded distribution profile and the reference distribution profile, that is, how the distribution profile and the reference distribution profile differ. If it is determined that the recorded distribution profile corresponds to the reference distribution profile, no further investigation is necessary.
[0023] Preferably, the evaluation unit is set up to determine differences between the recorded distribution profile and the reference distribution profile.
[0024] Preferably, the reference distribution profile comprises at least one reference characteristic, preferably several reference characteristics, and more preferably one reference characteristic for each characteristic. Preferably, each reference characteristic is assigned to and / or assignable to a characteristic.
[0025] According to a preferred embodiment, the evaluation unit may be configured to determine that the detected distribution profile corresponds to the reference distribution profile if at least one characteristic of the detected distribution profile differs from the reference characteristic by no more than one tolerance value, preferably, and / or lies within a tolerance range, preferably with respect to the reference distribution profile, more preferably with respect to a reference characteristic or a value of the reference characteristic. It is particularly preferred that all characteristics of the detected distribution profile differ by no more than the respective tolerance value and / or lie within the tolerance range.
[0026] According to a preferred embodiment, the evaluation unit may be configured to determine that the detected distribution profile does not correspond to the reference distribution profile if at least one characteristic of the detected distribution profile differs by at least one tolerance value and / or lies outside a tolerance range, preferably with respect to the reference distribution profile, more preferably with respect to a reference characteristic or a value of the reference characteristic. It may also be preferred that if at least one characteristic of the detected distribution profile differs from the reference characteristic, the detected distribution profile does not correspond to the reference distribution profile. Alternatively, it may be provided that the detected distribution profile does not correspond to the reference distribution profile if at least two characteristics of the detected distribution profile do not correspond to the associated reference characteristic.
[0027] If at least one characteristic of the recorded distribution profile differs from the reference characteristic by no more than the tolerance value and / or lies within the tolerance range of the reference characteristic around the reference characteristic, the distribution profile corresponds to the reference distribution profile.
[0028] The reference characteristic, or the value of the reference characteristic, can be understood as a target value of the characteristic.
[0029] Preferably, each (reference) characteristic is assigned and / or assignable a tolerance value and / or tolerance range.
[0030] Preferably, at least one tolerance value and / or tolerance range is variable. Preferably, each characteristic can be assigned its own tolerance value and / or tolerance range. Preferably, the respective tolerance value and / or tolerance range is selected from a range of 0% to 30%, preferably 5% to 20%, and particularly preferably 5% to 15% of the value of the at least one reference characteristic. It is also conceivable that the tolerance value and / or tolerance range each have a specific value, which is preferably dependent on the at least one characteristic. It is also conceivable that the tolerance value and / or tolerance range are asymmetrical, meaning that an upper tolerance value (i.e., a higher absolute value) differs from a lower tolerance value of the characteristic.
[0031] According to a preferred embodiment, the detection device may be a non-contact detection device. Preferably, the non-contact detection device may be an optical, electromagnetic, and / or similar detection device.
[0032] Preferably, the capture device is an image recording device, for example, a camera, preferably any type of camera. Preferably, the camera can be a 3D camera, a high-speed camera, a time-of-flight (TOF) camera, and / or the like. Preferably, the capture device is configured to record at least one image, at least one series of images, and / or a video.
[0033] Preferably, at least one characteristic of the distribution profile can be detected by the detection device. Preferably, the detection device is configured to detect at least one characteristic, a selection of characteristics, and / or every characteristic of the distribution profile. Preferably, the characteristics to be detected are predefined and / or predefined.
[0034] Preferably, the evaluation unit is configured to perform an evaluation of the distribution profile recorded by the recording device, preferably depending on at least one characteristic of the recorded distribution profile.
[0035] According to a preferred embodiment, the evaluation unit may be configured to compare the detected distribution profile with a reference distribution profile. Preferably, the evaluation unit is configured to determine, based on this comparison, whether the detected distribution profile corresponds to the reference distribution profile.
[0036] Preferably, the evaluation unit is configured to evaluate the recorded distribution profile with respect to at least one characteristic. Preferably, the evaluation unit is configured to determine at least one characteristic of the recorded distribution profile from the recorded distribution profile itself. This means, for example, that at least one characteristic of the recorded distribution profile can be determined from the recorded distribution profile using the evaluation unit, and that it can be determined using the evaluation unit whether the distribution profile, or at least one characteristic of the distribution profile, corresponds to the reference distribution profile or at least one reference characteristic.
[0037] Preferably, the evaluation unit is configured to examine at least one, a selection, or all characteristics of the recorded distribution profile for differences. This can be advantageous because each characteristic can influence the correct distribution of the material being distributed.
[0038] The characteristics to be examined can be predefined and / or predefinable by the evaluation unit. Alternatively or cumulatively, the evaluation unit can be configured to select the characteristics to be examined, preferably automatically and / or depending on a corresponding selection signal, for example, user input, a signal received from another component, e.g., a control unit, preferably a higher-level control unit, and / or the like.
[0039] Preferably, the evaluation unit is configured to be connected, at least via signals, to a storage unit, wherein at least one reference distribution profile, preferably comprising at least one reference characteristic, is stored in the storage unit. The storage unit can be part of the evaluation unit and / or be arranged outside the evaluation unit and be connected to the evaluation unit via signals.
[0040] Preferably, the evaluation unit comprises at least one processing unit configured to determine whether the recorded distribution profile corresponds to the reference distribution profile. Preferably, the processing unit of the evaluation unit is configured to perform all arithmetic operations of the evaluation unit. For example, this could include comparing a recorded distribution profile with a reference distribution profile, although this is not an exhaustive list. Preferably, the processing unit is configured to access the storage unit. The processing unit may also be configured to execute appropriate software.
[0041] Preferably, the evaluation unit can be configured to compare the recorded distribution profile with the reference distribution profile by means of difference calculation, quotient calculation, and / or other image analysis methods. Preferably, the evaluation unit is configured to compare at least one characteristic of the recorded distribution profile with the corresponding at least one reference characteristic of the reference distribution profile, in particular the value(s) of the at least one characteristic of the recorded distribution profile with the value(s) of the corresponding at least one reference characteristic of the reference distribution profile.
[0042] Preferably, the evaluation unit is configured to perform segmentation of the data from the acquisition device, particularly with regard to the acquired distribution profile. Preferably, the evaluation unit is configured to perform segmentation of an image using the acquired at least one distribution profile. Preferably, the at least one distribution profile can be recognized as an object in the image and thus preferably separated from the rest of the image. Segmentation is particularly advantageous when several distribution profiles are acquired simultaneously and several distribution profiles are, for example, depicted in one image. Preferably, the evaluation unit is configured to perform segmentation depending on the number of acquired distribution profiles and / or the at least one characteristic. Segmentation can reduce the required computing power and processing time.In particular, the evaluation unit can be configured to perform a classification of the segments. That is, it can identify whether they are indeed distribution profiles.
[0043] Preferably, the evaluation unit comprises an artificial intelligence unit with an artificial intelligence. Preferably, the artificial intelligence unit is configured, preferably using the artificial intelligence: - to adjust the reference distribution profile based on recorded distribution profiles; and / or - to create the reference distribution profile based on training data; and / or - to determine, based on the comparison, whether the distribution profile corresponds to the reference distribution profile; and / or - to determine the differences between the recorded distribution profile and the reference distribution profile.
[0044] Preferably, the artificial intelligence unit comprises appropriate software and / or hardware. Preferably, the software is executable by the computing unit, which preferably comprises at least one CPU, at least one GPU, at least one NPU, and / or the like. Preferably, the artificial intelligence is machine learning, preferably deep learning.
[0045] The input data for the artificial intelligence unit can be the recorded distribution profiles. The output data can indicate whether the recorded distribution profile corresponds to the reference distribution profile or not, and / or indicate which characteristic of at least one characteristic differs. Preferably, the artificial intelligence unit can be configured to perform the segmentation and, if necessary, the classification.
[0046] Preferably, the evaluation unit is equipped with the artificial intelligence unit to perform the comparison between the recorded distribution profile and the reference distribution profile.
[0047] Preferably, the evaluation unit is configured with the artificial intelligence unit to adjust the reference distribution profile based on acquired distribution profiles. Each acquired distribution profile can be determined by comparison with the reference distribution profile as either conforming to or deviating from it. Based on the insights gained, the artificial intelligence unit can be configured to adjust the reference distribution profile. Preferably, the values of the reference characteristics, the tolerance values, the tolerance ranges, and / or the like can be adjusted by means of the artificial intelligence unit. In particular, the respective values can be further refined. Preferably, the corresponding detection of whether the acquired distribution profile corresponds to the reference distribution profile can be improved.
[0048] Preferably, the artificial intelligence unit or the artificial intelligence itself can be trained using training data to obtain a faster and more precise assessment of a recorded distribution profile. Preferably, the artificial intelligence unit can be configured to create the reference distribution profile based on the training data and / or to adapt an existing reference distribution profile.
[0049] The training data can, for example, consist of images of distribution profiles under specific conditions, such as pressure, the type of distribution unit, and / or the like. Preferably, the training data comprises distribution profiles from distribution units that function flawlessly. Preferably, the training data consists of a multitude of distribution profiles from a multitude of identical distribution units. This can be advantageous because, despite the identical construction of the distribution units, differences in the distribution profiles can occur. The training data can include the distribution profiles for flawlessly functioning distribution units that are obtained and / or expected under the set pressure and / or the corresponding configuration of the distribution unit.
[0050] Preferably, the artificial intelligence unit can be configured to determine expected distribution profiles, preferably based on the parameters. Preferably, the set pressure and / or similar parameters can be taken into account. Preferably, the expected distribution profiles can be stored as reference distribution profiles. It is preferably provided that the reference distribution profiles have been created using the artificial intelligence unit.
[0051] According to a further preferred embodiment, it can be provided that, by means of a user input, it can be specified and / or automatically specified with respect to which at least one characteristic the recorded distribution profile is to be examined by the evaluation device, that is, preferably, with respect to which characteristic the evaluation unit is to determine whether it corresponds to the reference characteristic or not.
[0052] Preferably, the system includes an input console configured to allow user input. Preferably, the evaluation unit is configured to examine at least one characteristic of the recorded distribution profile based on the user input. Alternatively or cumulatively, the evaluation unit may be predefined as to which at least one characteristic to examine. Preferably, by examining at least one characteristic of the recorded distribution profile, it can be determined whether and which characteristic deviates. Preferably, the cause of a distribution profile that does not correspond to the reference distribution profile can be narrowed down by identifying deviating characteristics. Preferably, the predefined characteristic(s) can be modified and / or supplemented as desired.
[0053] According to a preferred embodiment, a lighting device may be provided which is configured to illuminate the distribution profile, in particular the distributed material which emerges from at least one distribution unit.
[0054] By illuminating the distribution profile, the corresponding reflection can be used to better visualize the profile, as it stands out more clearly from the background. Preferably, the lighting device is configured to illuminate the material being distributed in all distribution units. Preferably, the lighting device comprises at least one lighting unit, each assigned to one or more distribution units. The lighting device can preferably be a continuous light, a continuously switched-off light, a strobe light, and / or the like. Preferably, the lighting device can be LED-based, meaning it comprises one or more LEDs.
[0055] Preferably, the control unit is configured to control the lighting device. The control is preferably such that detection by the detection device is possible when the material being distributed within the distribution profile is illuminated.
[0056] According to a further preferred embodiment, the detection device may include a detection area, preferably a variable detection area. Variable may mean, for example, with regard to width, height, position, focal point, magnification, and / or the like. Preferably, the detection area may be dependent on at least one characteristic to be examined.
[0057] Preferably, the detection device is configured to be displaceable and / or rotatable relative to the distribution rod. Preferably, a displacement device is provided which is configured to mount the detection device so that it is displaceable and / or rotatable relative to the distribution rod.
[0058] By moving and / or rotating the detection device, distribution profiles from several distribution units can be detected, preferably one after the other.
[0059] According to a further preferred embodiment, the detection device can be configured to continuously, discontinuously, at predetermined and / or predefinable times, and / or based on user input. A user input can transmit a request to the detection device to detect a distribution profile, preferably at least one specific distribution profile. Preferably, the detection device detects the distribution profile, preferably precisely when the distribution unit is active, i.e., when material is being dispensed by the distribution unit.
[0060] Preferably, the data acquisition device and / or the evaluation unit is configured to check whether the distribution unit is and / or was active during the acquisition of the distribution profile of the at least one distribution unit. If the distribution unit is or was not active, no acquisition of the at least one distribution profile or the acquisition of the at least one characteristic can be / could have been carried out.
[0061] Preferably, the detection device may be configured to recognize whether the respective distribution unit is active. Preferably, the detection device may perform a detection of the distribution profile when the distribution unit is active. Preferably, the detection device is configured to perform a detection of the distribution profile when the distribution unit is active.
[0062] Preferably, the system includes a control unit configured at least for controlling the distribution units. This means that the distribution units can be activated or deactivated by corresponding control signals from the control unit. Preferably, the control unit can transmit the control signals for the distribution units to the acquisition device and / or, depending on the control signals from the distribution units, control the acquisition device to perform the acquisition of the distribution profile(s).
[0063] According to a further preferred embodiment, it can be provided that at least a subset of the distribution units is assigned an identification feature, preferably a machine-readable identification feature. Preferably, each distribution unit is assigned a corresponding identification feature.
[0064] Preferably, the identification feature can be attached to the distribution unit and / or near the distribution unit on the distribution boom, preferably in such a way that the identification feature is also detectable by the detection device. This facilitates the easier location of a distribution unit with a faulty distribution profile. Preferably, the identification feature is a code, a number, a pattern, and / or the like, and any combination thereof. Preferably, it can be a barcode, QR code, and / or the like. It is preferably conceivable that the identification feature is attached to the distribution boom and / or to the respective distribution unit itself at the respective positions of the distribution unit.
[0065] According to a further preferred embodiment, a warning unit may be provided which is configured to issue a warning, issue a recommendation for action, and / or initiate an automatic reaction if the detected distribution profile is faulty. Preferably, the warning may be visual, audible, and / or haptic. A visual warning may preferably be displayed on a screen, by light elements, for example in a driver's cab or the like, and / or the like. An audible warning may be a tone and / or a sequence of tones. A haptic warning may be a vibration of a component, for example, a component necessary for the operation of the system, such as a console, a control stick, and / or the like.
[0066] Preferably, the warning unit can be part of the evaluation unit and / or be connected to the evaluation unit via a signal connection. Preferably, the evaluation unit is configured to output the warning message to the warning unit and / or to transmit it to a corresponding device that is configured to output the warning message.
[0067] A recommended action, for example through a display, a tone, or a sequence of tones, can indicate to the user that they should perform a specific action, such as stopping the system, performing a repair soon, and / or the like. Preferably, the evaluation unit is configured to output the recommended action and / or transmit it to a corresponding device that is configured to output the recommended action.
[0068] An automatic response could be that the system is automatically stopped if a faulty distribution profile is present. Preferably, the evaluation unit and / or the warning unit is configured to output a corresponding signal, for example to the control unit, and to cause the control unit to execute the automatic response. This preferably means controlling components accordingly.
[0069] The underlying problem is also solved by a method for distributing agricultural material on an agricultural area, preferably by means of a system according to an embodiment, at least comprising the following process steps: a) Recording at least one distribution profile of at least one distribution unit by means of a recording device; b) Determine, using an evaluation unit, whether the recorded distribution profile is faulty.
[0070] The underlying problem is also solved by an agricultural machine characterized by having a system according to an embodiment and / or being configured to carry out a method according to an embodiment. The agricultural machine could be, for example, a field sprayer, a pneumatic fertilizer spreader, and / or the like.
[0071] In the context of the invention, the individual steps of the process can be carried out in a defined sequence; however, it is also conceivable that the steps of the process can be carried out in any sequence. An arbitrary change between the process steps is also conceivable. Furthermore, the process can be extended by adding further process steps.
[0072] Preferably, the steps of the procedure are carried out continuously and / or at intervals.
[0073] The invention further comprises a method for generating training data for an artificial intelligence unit, comprising the following method steps: - Providing: - a large number of identical distribution units; or - several groups of different types of distribution units, each group comprising a multitude of identical distribution units, for a system and / or an agricultural machine; - Operating the distribution units; - Recording distribution profiles of each distribution unit, preferably by means of a recording device.
[0074] The operation of the distribution units can preferably be such that the multitude of identical distribution units are initially operated under the same conditions. Once all distribution profiles of each identical distribution unit have been recorded, the conditions under which the distribution profiles are operated are changed. The multitude of identical distribution units is then operated under the modified, but identical, conditions.
[0075] Preferably, the artificial intelligence unit can be configured, preferably based on the training data, to recognize which conditions exist during operation.
[0076] Preferably, it may be provided that the associated characteristics are recorded for each of the recorded distribution profiles.
[0077] Preferably, the distribution units may be operated under certain conditions, such as an applied pressure, the type and / or composition of the distributed material, and / or the like.
[0078] The invention also includes the use of a trained artificial intelligence unit, which is trained with the training data, to - to create a reference distribution profile based on the training data; and / or - to adjust the reference distribution profile based on recorded distribution profiles; and / or - to perform a comparison between the recorded distribution profile and the reference distribution profile; and / or - To determine differences between the recorded distribution profile and the reference distribution profile.
[0079] Preferably, the training data, i.e., the recorded distribution profiles, can be stored in the storage device. Alternatively, it can be provided that the reference distribution profile is generated from the training data and stored in the storage device.
[0080] The control unit includes, for example, a computer unit, an on-board computer, and / or similar components, and also comprises a control and / or regulation circuit, in particular an electrical control and / or regulation circuit, wherein the control and / or regulation circuit is suitably designed for electrical signal and / or command transmission. This signal and / or command transmission can also be wireless (e.g., via WLAN).
[0081] In the context of the invention, the term "control unit" encompasses, in particular, the entirety of components for signal and / or command transmission. Accordingly, this also includes computer units, CPUs, and / or the like. Likewise, it also includes control devices integrated into the respective sensors, sensor units, or sensor arrangements. It should also be noted that the signals and / or data from the sensors, measuring devices, detection devices, and / or the like can each be used as feedback for a control variable.
[0082] It should be noted that the terms "control," "regulate," "control unit," and "regulation unit" can refer to electronic controls or regulators that, depending on their design, can perform control and / or regulation tasks. Even though the term "control" is used here, it can also appropriately encompass "regulation." Likewise, the use of the term "regulation" can also imply "control."
[0083] Regarding the advantages and embodiments of the method according to the invention, reference is made to the advantages and embodiments of the agricultural distribution machine according to the invention.
[0084] To avoid repetition, it should be noted that the embodiments and features according to the invention can be combined in any way and freely with the system, the method, and / or the agricultural machine. Accordingly, all embodiments and features according to the invention are disclosed and claimable for the system, the method, the agricultural machine, the method for generating the training data, and the use of the artificial intelligence unit.
[0085] For the purposes of the application, features disclosed in conjunction with other features may also be considered disclosed on their own. Features linked by "and / or" are to be understood as disclosed both on their own and in combination with the other features.
[0086] Further details and advantages of the invention are described below with reference to the accompanying drawings. The relative sizes of the individual elements in the figures do not always correspond to the actual relative sizes, as some shapes are simplified and others are enlarged for better illustration in relation to other elements. The figures show: Fig. 1 system or agricultural machine according to one embodiment; Fig. 2. Correct distribution profile of a distribution unit; Fig. 3 faulty distribution profile of a distribution unit; Fig. 4 System in a schematic representation according to one embodiment; Fig. 5 methods according to one embodiment.
[0087] The in the Fig. The embodiments shown in Figures 1 to 5 are at least partially identical, so that similar or identical parts are provided with the same reference numerals and, to avoid repetition, reference is also made to the description of the other embodiments or figures for their explanation.
[0088] Fig. Figure 1 shows a system 1 or an agricultural machine 100, which is exemplified as a field sprayer. This embodiment is merely exemplary and not to be understood as exhaustive.
[0089] The field sprayer includes, among other things, a tank 11 in which the distribution material 2, in particular spray liquid, can be stored. Furthermore, a conveying device for pumping the spray liquid from the tank 11 to distribution units 5 may be provided, although the conveying device is not shown.
[0090] A distribution linkage 3 is further shown, which has a plurality of distribution sections 4. The distribution sections 4 can also be referred to as booms and are movable relative to each other. Preferably, joints 14 are provided, which are configured so that two distribution sections 4 can move relative to each other, in particular, be foldable relative to each other. A central distribution section 4.1 can also be provided, which is coupled to a support frame 12. The support frame 12 is configured to move the distribution linkage 3 vertically. Preferably, the support frame 12 is designed as a parallelogram arrangement. Preferably, the support frame 12 is coupled to a chassis frame 15, wherein wheels 13 for support against an agricultural surface F are also coupled to the chassis frame 15.
[0091] Several distribution units 5 are arranged on the distribution rod 3 or on the distribution sections 4, 4.1, preferably evenly spaced. Each distribution unit 5 is designed to dispense distribution material 2 with a distribution profile 6.
[0092] In the Fig. Figure 1 shows only some of the distribution profiles 5. The distribution units 5 can also be referred to as spray nozzles. A characteristic can be a value that specifies the distribution profile 6. Preferably, at least one characteristic of the distribution profile 6 is one from the group consisting of: opening angle, shape, uniformity of distribution, size, distribution pattern on an agricultural area, material spectrum, average particle size of the material being distributed, exit velocity of the material from the distribution unit, quantity of material being distributed, and any combination thereof, and / or the like. Preferably, the distribution profile can have a spatial distribution, which can be referred to, for example, as a distribution fan, distribution cone, spray fan, and / or the like. The spatial distribution can be described by one or more characteristics.
[0093] Furthermore, a detection device 7 with a detection area 16 is shown schematically, whereby the detection device 7 with the detection area 16 is only shown schematically.
[0094] The detection device 7 is configured to detect the distribution profile 5 of at least one distribution unit 5. Preferably, the detection device 7 is configured to transmit sensor data, i.e., the detected distribution profile 6 or the characteristics of the distribution profile 6, to an evaluation unit 8.
[0095] The evaluation unit 8 is configured to determine whether the distribution profile 6 corresponds to a reference distribution profile. The evaluation unit 8 is preferably configured to receive the distribution profile 6 acquired by the acquisition device. The evaluation unit 8 is configured to analyze and / or process the acquired distribution profile 6.
[0096] Preferably, one or more characteristics can be provided by which the distribution profile 6 is characterized. Preferably, the characteristics can be predefined and / or predefinable.
[0097] In the Fig. Figure 2 shows a correct distribution profile 6 of a distribution unit 2, with an opening angle 17, a distribution material spectrum 18 (here a droplet spectrum), a uniform distribution, and a specific shape, to name just a few possible characteristics. The quantity of the distribution material can also be considered a characteristic. A correct distribution profile 6 is understood to be one that corresponds to a reference distribution profile.
[0098] In the Fig. Figure 3 shows a faulty distribution profile 6, i.e., a distribution profile that does not correspond to the reference distribution profile, for example, missing several droplet jets. Other jets are significantly attenuated. Therefore, the shape, the distribution with respect to intensity and homogeneity, as well as the opening angle 17 and the distribution pattern are all altered.
[0099] Furthermore, an identification feature 11 is shown, which is attached to the distribution unit 5. The distribution unit 5 can be uniquely identified by means of the identification feature 10.
[0100] Likewise, a lighting device 9 is provided, which is designed to illuminate the distribution profile 6 in order to enable better detection by the detection device 7.
[0101] In the Fig. Figure 4 schematically illustrates system 1 according to one embodiment.
[0102] The distribution profile 6 of a distribution unit 5 is recorded by the acquisition device 7 and transmitted to the evaluation unit 8. The evaluation unit 8 is configured to determine whether at least one characteristic corresponds to the associated reference characteristic, preferably by means of a processing unit 20. The classification is performed by comparing the recorded distribution profile 6 with a reference distribution profile stored in a storage unit 19. The evaluation unit 8 is configured to access the storage unit 19. Preferably, a warning unit 21, in this case in the form of a display unit 21, can be provided, which is configured to indicate whether and / or which characteristic(s) are classified as faulty or correct. The evaluation unit 8 and the display unit 21 are configured to exchange data with each other.The display unit 21 is preferably configured to function as an input unit, for example for touches by a person and / or the like.
[0103] In the Fig. Figure 5 schematically shows a method for distributing material. In the present embodiment, the method comprises providing a system 1 with a distribution unit 5, a detection device 7, and an evaluation unit 8, preferably providing a system 1 according to a preferred embodiment (S1). In a further step (S2), the distribution profile of the at least one distribution unit is detected by means of the detection device. In a further step (S3), the evaluation unit determines whether the detected distribution profile corresponds to a reference distribution profile.
[0104] Although the invention has been described with reference to specific embodiments, it is apparent to a person skilled in the art that various modifications can be made and equivalents used as substitutes without departing from the scope of the invention. Furthermore, many modifications can be made without departing from the relevant scope. Consequently, the invention is not intended to be limited to the disclosed embodiments but is intended to encompass all embodiments falling within the scope of the appended claims. In particular, the invention also claims protection for the subject matter and features of the dependent claims independently of the referenced claims. Reference symbol list 1 system 2 Distributed goods 3 distribution rods 4 Distribution section 5 distribution unit 6 Distribution profile 7. Recording device 8 evaluation units 9 Lighting equipment 10 Identification feature 11 Tank 12 support frames 13 wheel 14 joint 15 chassis frames 16. Detection area 17 opening angles 18 Distribution spectrum 19 storage units 20 computing units 21 Warning unit / Display unit 100 agricultural machines F agricultural area
Citation Information
Patent Citations
Agricultural spreading machine and method for examining a spreading element
DE102018125152A1