Efficient spot spraying of agricultural chemicals

Abstract

Described herein is a method of operation of a control system for spraying agricultural chemicals using a spray assembly. The method includes identifying a target object and proximate avoidance objects, and then defining a pattern of spot sprays by placement and open duration of the spot sprays such that a maximum non-zero dose applied to the proximate avoidance objects is not exceeded and at least a minimum dose per unit area is applied to a predefined proportion of the target object. Each spot spray is defined by a non-uniform two-dimensional liquid spatial distribution that is modified by distance from the nozzle to the target object, forward speed of the spray assembly, and open duration of the nozzle. After the pattern is defined, the method includes generating control signals for an array of nozzles according to the pattern of spot sprays, and outputting the control signals to the spray assembly.

Description

Technical Field

[0001] This invention relates to the control of spraying equipment for spot spraying of plants with agricultural chemicals.

[0002] background Agricultural chemicals, whether promoting growth (e.g., fertilizers), inhibiting growth (e.g., herbicides), or preventing disease or pests (fungicides, insecticides, etc.), are typically applied to plants in liquid form using a sprayer. Agricultural chemicals are sprayed through nozzles that can be mounted on a spray bar. The spray bar can be mounted on a vehicle (e.g., a tractor or robot) or on a device towed by a vehicle. Continuous spraying, or broadcast spraying, involves spraying agricultural chemicals in a continuous manner, with the nozzles operating continuously. In contrast, spot spraying applies droplets to specific and predetermined locations using a valve (e.g., an electromechanical control valve) that can quickly open and close the flow of agricultural chemicals.

[0003] Continuous spraying is generally inefficient because agrochemicals are sprayed where they are not needed (e.g., on bare soil) and increases costs. It also increases chemical residues in the soil, which can have various effects, including damaging biodiversity and increasing the likelihood of phytotoxicity to the sprayed crops, leading to yield losses. Targeted spraying can significantly reduce the amount of agrochemicals applied. This improves efficiency (because agrochemicals are applied only where needed), reduces environmental impact (e.g., fewer chemical residues in soil and water, and reduced carbon emissions due to reduced manufacturing and transportation of liquid agrochemicals), reduces water use, and increases yields.

[0004] Overview This overview is provided to introduce, in a simplified form, some selected concepts that are further described in the detailed description below. This overview is not intended to identify key or essential features of the claimed subject matter, nor is it intended to help determine the scope of the claimed subject matter.

[0005] This document describes an operational method for a control system used in spraying agricultural chemicals using a spray assembly. The method includes identifying target objects and adjacent avoidance objects, and then defining a pattern for the targeted spray by the placement and opening duration of the targeted spray, such that a maximum non-zero dose applied to adjacent avoidance objects is not exceeded, and a minimum dose per unit area is applied to the target object at a predefined proportion. Each targeted spray is defined by a non-uniform two-dimensional liquid spatial distribution, which is modified by the distance from the nozzle to the target object, the forward velocity of the spray assembly, and the opening duration of the nozzle. After defining the pattern, the method includes generating control signals for a nozzle array based on the pattern of the targeted spray and outputting the control signals to the spray assembly.

[0006] A first aspect provides an operating method for a targeted spray control system for spraying agricultural chemicals onto a target object using a spray assembly and minimizing spraying onto avoidable objects, the spray assembly including a nozzle array, wherein the opening time and opening duration of the nozzles can be individually controlled, and the method includes: (i) identifying the target object and adjacent avoidable objects; (ii) defining a pattern of targeted spray by the placement and opening duration of the targeted spray, such that a maximum non-zero dose applied to adjacent avoidable objects is not exceeded, and a minimum dose per unit area is applied to the target object at a predefined proportion, wherein each targeted spray is defined by a non-uniform two-dimensional liquid space distribution modified by the distance from the nozzle to the target object, the forward speed of the spray assembly, and the opening duration of the nozzle; (iii) generating a control signal for the nozzle array based on the pattern of the targeted spray; and (vi) outputting the control signal to the spray assembly.

[0007] Defining the pattern of the spot spray may include, for each target object: placing a first spot spray as close as possible to the adjacent avoidance object without exceeding the maximum non-zero dose on the adjacent avoidance object; placing one or more additional spot sprays as close as possible to the adjacent avoidance object to form a continuous spray area with at least a minimum dose on a portion of the target object, without exceeding the maximum non-zero dose on the adjacent avoidance object; and placing one or more additional spot sprays such that the continuous spray area with at least a minimum dose extends over a predefined proportion of the target object if the portion of the target object covered by the continuous spray area is smaller than a predefined proportion of the target object.

[0008] Defining a pattern for targeted spraying may include: determining the spray size and stride such that, when the targeted sprays are staggered, no area receives a dose less than a minimum dose; generating multiple resulting patterns, each corresponding to a different candidate starting position among multiple candidate starting positions for the targeted spray array; selecting from the multiple resulting patterns the resulting pattern that minimizes waste of agrochemicals, wherein each resulting pattern is generated by: for each target object, generating an array portion by: positioning the targeted spray array across the entire spray window with the determined size and stride, starting from a candidate starting position; removing all targeted sprays from the array that do not reach the target object; removing targeted sprays until the dose received by adjacent avoidable objects is less than a maximum non-zero dose; removing any targeted sprays that are unnecessary for achieving the minimum dose at a predefined proportion for the target object; and combining the array portions for each target object.

[0009] Defining a pattern for targeted spraying may include: determining the spray size and stride such that when the targeted sprays are staggered, no area receives a dose less than a minimum dose; generating multiple array portions for each target object, each array portion corresponding to a different candidate starting position among multiple candidate starting positions of the targeted spray array; and selecting from the multiple array portions the array portion that minimizes waste of agrochemicals, wherein each array portion is generated by: positioning the targeted spray array at the determined size and stride across the entire spray window, starting from the candidate starting position; removing all targeted sprays from the array that do not reach the target object; removing targeted sprays until the dose received by adjacent avoidable objects is less than a maximum non-zero dose; and removing any targeted sprays that are unnecessary for achieving a minimum dose at a predefined proportion on the target object.

[0010] The maximum non-zero dose can be the maximum non-zero dose per unit area.

[0011] The method may further include: determining whether there is more than one target object approaching the avoidance object; and in response to determining that there is more than one target object approaching the avoidance object, dividing the maximum non-zero total dose for the avoidance object into an allocation amount for each target object, wherein the allocation amount is the maximum non-zero dose used when defining the pattern for the target object.

[0012] The method may further include: determining whether there is more than one target object approaching the avoidance object; in response to determining that there is more than one target object approaching the avoidance object, determining whether the maximum non-zero total dose for the avoidance object is given a higher dose per unit area than the local maximum non-zero dose per unit area for the avoidance object; and in response to determining that the maximum non-zero total dose for the avoidance object is not given a higher dose per unit area than the local maximum non-zero dose per unit area for the avoidance object, dividing the maximum non-zero total dose for the avoidance object into an allocation amount for each target object, wherein the allocation amount is the maximum non-zero dose used when defining the pattern for the target object.

[0013] Defining a pattern for targeted spraying may include: determining the spray size and stride such that, when the targeted sprays are staggered, no area receives a dose less than a minimum dose; generating multiple resulting patterns, each corresponding to a different candidate starting position among multiple candidate starting positions for the targeted spray array; and selecting from the multiple resulting patterns the resulting pattern that minimizes waste of agrochemicals, wherein each resulting pattern is generated by: positioning the targeted spray array across the entire spray window with the determined size and stride, starting from a candidate starting position; removing all targeted sprays from the array that do not reach any target object; removing targeted sprays until each adjacent avoidable object receives a dose less than a maximum non-zero dose; removing any targeted sprays that are unnecessary for reaching a minimum dose at a predefined proportion for any target object; and combining array portions for each target object.

[0014] The predefined proportion of the target object can be less than the entire target object, and the method may further include: identifying a portion of the target object that has not received the minimum dose; and updating the pattern of the spot spray by adding one or more additional spot sprays to that portion of the target object.

[0015] The predefined proportion of the target object can be the entire target object.

[0016] The two-dimensional liquid spatial distribution of a fixed-point spray can be determined based on the static two-dimensional liquid spatial distribution of the nozzle and the motion data of the received spray assembly.

[0017] The two-dimensional liquid spatial distribution of a fixed-point spray can be determined based on the static two-dimensional liquid spatial distribution of the nozzle and the detected distance between the nozzle and the target object.

[0018] The two-dimensional liquid spatial distribution of a fixed-point spray can be determined based on the static two-dimensional liquid spatial distribution of the nozzle and the nozzle opening duration.

[0019] The two-dimensional liquid spatial distribution of a fixed-point spray can be defined in a lookup table.

[0020] The two-dimensional liquid spatial distribution of a fixed-point spray can be defined using a mathematically defined distribution.

[0021] The method may further include: adjusting the nozzle opening duration in response to changes in the nozzle advance speed during spraying to maintain a constant two-dimensional liquid spatial distribution for the point spray.

[0022] The method may also include: adjusting a predefined minimum dose based on the size or type of the target object.

[0023] The second aspect provides a computer program that includes instructions that, when executed by a computer, cause the computer to perform any of the methods described above (including any combination of the features described above).

[0024] The third aspect provides a computer-readable medium having the aforementioned computer program stored thereon.

[0025] A fourth aspect provides a targeted spraying control system for spraying agricultural chemicals using a spray assembly comprising a nozzle array, wherein the opening time and opening duration of the nozzles can be individually controlled, and the targeted spraying control system comprising: a processor; one or more interfaces configured to receive target object data and output control signals to the spray assembly; and a memory arranged to store a computer program that, when executed by the processor, causes the control system to: (i) identify target objects and adjacent avoidance objects; (ii) define a pattern of targeted spraying by the placement and opening duration of the targeted sprays such that a maximum non-zero dose applied to adjacent avoidance objects is not exceeded, and a minimum dose per unit area is applied to the target objects at a predefined proportion, wherein each targeted spray is defined by a non-uniform two-dimensional liquid space distribution modified by the distance from the nozzle to the target object, the forward speed of the spray assembly, and the opening duration of the nozzles; (iii) generate control signals for the nozzle array based on the pattern of the targeted sprays; and (vi) output the control signals to the spray assembly.

[0026] The methods described herein can be executed by software in a machine-readable form on a tangible storage medium, for example, in the form of a computer program comprising computer program code means adapted to perform all steps of any of the methods described herein when the program is run on a computer, and wherein the computer program can be contained on a computer-readable medium. Examples of tangible (or non-transitory) storage media include disks, thumb drives, memory cards, etc., and do not include propagation signals. The software may be adapted to execute on a parallel or serial processor, such that the method steps can be executed in any suitable order or simultaneously.

[0027] This acknowledges that firmware and software can be valuable, separately tradable commodities. It is intended to include software that runs on or controls “non-intelligent” or standard hardware to perform desired functions. It is also intended to include software that “describes” or defines hardware configurations, such as HDL (Hardware Description Language) software used to design silicon chips or to configure general-purpose programmable chips to perform desired functions.

[0028] The embodiments described below are not limited to implementations that address any or all the drawbacks of known methods for controlling the spraying of agricultural chemicals.

[0029] Preferred features may be combined where appropriate, as will be apparent to those skilled in the art, and may be combined with any aspect of the invention. Brief description of the attached diagram Embodiments of the present invention will be described by way of example with reference to the following accompanying drawings, in which: Figure 1 This is a schematic diagram of the first example of a fixed-point spraying system; Figure 2 This is a schematic diagram of the second example of a fixed-point spraying system; Figure 3 This is a schematic diagram of a dual-spray bar arrangement; Figure 4 Two different examples of one-dimensional (1D) point spray dose profiles are shown; Figure 5 An example shape of a targeted spray is shown; Figure 6 An example representation of the 2D liquid spatial distribution of a fixed-point spray is shown in the form of a lookup table, as well as the 1D liquid spatial distribution along the central transverse and longitudinal axes through the fixed-point spray; Figure 7 Two examples of the resulting 2D liquid spatial distribution of a fixed-point spray as a result of different opening durations and forward movement are shown; Figure 8 A point spray control system (such as) is shown. Figure 1 Example operation method of the fixed-point spray control system 102 shown.

[0031] Figure 9 The generation of [something] is shown. Figure 8 The first example method for using the point spray pattern in the method; Figure 10 and Figure 11 It shows the use Figure 9 The method generates a graphical representation of the fixed-point spray pattern; Figure 12A The generation of [something] is shown. Figure 8The second example method for using the point spray pattern in the method; Figure 12B It shows Figure 12A Variations of the method; Figure 13A The generation of [something] is shown. Figure 8 The third example method for using the point spray pattern in the method; Figure 13B It shows Figure 13A Variations of the method; Figures 14-18 It shows the use Figure 13A The method generates a graphical representation of the fixed-point spray pattern; Figure 19 The generation of [something] is shown. Figure 8 The fourth example method for the point spray pattern used in the method; Figure 20 The generation of [something] is shown. Figure 8 The fifth example method for using the point spray pattern in the method; Figure 21 An example method for adjusting the minimum dose is shown; Figure 22 Modifications were shown. Figure 9 , Figure 12A , Figure 12B , Figure 13A , Figure 13B , Figure 19 and Figure 20 A method for generating a fixed-point spray pattern using any of the methods listed; Figure 23 The various components of an example targeted spray control system employing a computation-based device form are shown; and Figure 24 This is a schematic diagram showing overlapping spray windows.

[0032] Common reference figures are used throughout the figures to indicate similar features.

[0033] Detailed description Embodiments of the invention are described below by way of example only. These examples represent the best mode currently known to the applicant for carrying out the invention, although they are not the only ways in which this can be achieved. The description illustrates the function of the examples and the order of steps for constructing and operating the examples. However, the same or equivalent function and order may be accomplished by different examples.

[0034] As mentioned above, targeted spraying of agricultural chemicals can be significantly more effective than continuous spraying. However, for this to be effective, targeted spraying requires an efficient control system that translates the target spray area into a set of control signals for valves associated with the nozzles. Errors in this translation result in a mismatch between the target and actual spray areas, which can reduce efficiency (e.g., when the actual spray area is larger than the target spray area) and effectiveness (e.g., by not spraying portions of the target spray area). In many applications, the target spray area is determined in real time (e.g., using a camera system that scans the area of ​​the field before the spray bar passes through), so the available time to perform the translation is very short (e.g., less than 500 ms).

[0035] A crucial spraying parameter in agricultural chemical application is the control of dosage per unit area. Maintaining the highest level of spatial dosage uniformity is essential. In fact, insufficient dosage reduces chemical efficiency and can impair operations. Excessive dosage may exceed legal limits and be harmful to the environment, and / or cause damage to crops through phytotoxicity. Applying excessively high dosages per unit area to crops can also lead to localized scorching. Overdosing can also result in economic losses due to using more product than necessary.

[0036] Known control methods for spraying systems for agricultural chemicals rely on each nozzle providing uniform application within the spray pattern shape of the nozzle. In the case of continuous spraying using an array of nozzles mounted orthogonally to the direction of displacement of the application machine, good dosage uniformity is achieved by using nozzles with a uniform lateral distribution profile, having a relatively large jet divergence angle (typically 80 to 110 degrees) in the lateral direction, and typically spaced laterally at a distance less than their vertical distance to the target. This results in the droplet jets generated by adjacent nozzles largely overlapping and combining into a uniformly distributed, dense lateral cloud of droplets upon impact with the ground.

[0037] For spot spraying, uniform application of droplets is significantly more difficult for several reasons. First, because the area covered by a single spot spray must be as small as possible to achieve maximum spatial selectivity (e.g., to apply agricultural chemicals where needed rather than on adjacent plants), the nozzle jet divergence angle must be much smaller than that of continuous spraying (typically 15 to 25 degrees), which does not contribute to uniform dosage over a large area. Second, the nozzle-to-target distance must be as short as possible to ensure the highest possible spot placement accuracy. Therefore, when both the nozzle-to-target distance and the jet divergence angle are small, overlapping adjacent jets is a challenging task because the resulting required nozzle-to-nozzle lateral distance becomes extremely small. Such a high density of nozzles necessitates the use of low-flow nozzles to avoid over-application. Low-flow nozzles are difficult to manufacture and prone to clogging. Third, because the movement of the application machine can be rapid (e.g., several meters per second), the opening duration of the electromechanical valves associated with the nozzles must be very short to create short spots in the forward direction. This presents a challenge for valve design. Fourth, the distance from the nozzle to the target can vary during application due to the movement of the spray boom from the ground or irregular ground surfaces, or the height difference between the targets to be sprayed (e.g., in the case of varying plant height). This varying distance from the nozzle to the target makes it difficult to control the adjacent overlap ratio. When the overlap ratio is not well controlled, it is difficult to achieve uniformity through overlap based on a specific nozzle lateral dose profile, because the overlap can quickly change from a 2-jet overlap to a 3-jet overlap, for example, as the vertical distance increases.

[0038] All these factors make it difficult to apply a uniform dose of spray at a single point, whether for a single spray (i.e., when a single nozzle is activated) or for a larger spray that results in the cumulative effect of several adjacent nozzles. However, dose control remains a significant issue in targeted spraying equipment. Therefore, there is a strong need to develop methods for controlling the dose in targeted spraying equipment characterized by a dense array of nozzles with a small jet divergence angle and a small nozzle-to-target distance (typically 20 cm to 30 cm).

[0039] This paper describes an improved method for controlled spot spraying of agrochemicals. The method defines a spot spray pattern based on the detection locations of one or more target objects and one or more avoidance objects. Target objects are plants that need spraying, and avoidance objects are plants that do not need spraying. The spot spray pattern is defined such that a predefined non-zero maximum dose of agrochemicals targeting the avoidance objects (where the maximum dose can be a global dose or a local dose) is not exceeded, and the aim is to deliver as much (if not all) of a predefined minimum local dose as possible to the target objects, even if the target objects are adjacent to the avoidance objects. After defining the spot spray pattern, a control signal for the nozzle array in the spray assembly is generated based on the pattern and output to the spray assembly.

[0040] While in many applications of agrochemicals, dosage control ensures that the desired target object to be sprayed receives a minimum amount of product per unit area, several applications require the opposite dosage control: ensuring that a given object not to be sprayed (hereinafter referred to as the avoidable object) does not receive more than a given amount of product per unit area. This invention specifically addresses this second requirement (dosage control of the avoidable object) while remaining fully compatible with the first requirement (dosage control of the sprayable object).

[0041] The agrochemicals sprayed using the methods described herein can serve any purpose found in plant protection or fertilization; therefore, the plant that is a detected target or can form part of a detected target (where the target is a clump of plants) can be a crop or a weed. Similarly, the plant that is a detected avoidance target or can form part of a detected avoidance target (where the target is a clump of plants) can be a crop or a weed. Spraying an avoidance target is detrimental to the overall objective of the use of the agrochemical, for example, because it results in the crop being sprayed with herbicides or it results in the weeds being sprayed with fertilizers.

[0042] Examples of agricultural chemicals that can be sprayed using the methods described herein include fertilizers, herbicides, fungicides, insecticides, or other agricultural chemicals that promote or inhibit growth or prevent disease or pests. Other examples of agricultural chemicals that can be sprayed using the methods described herein include other liquids, such as hot water or hot oil for thermal weeding, or any other liquid used for weeding or crop care operations.

[0043] The method described herein can be implemented in real time as the spray assembly traverses the field and sprays detected target objects. As described in more detail below, the method can be implemented within each spray window, the size of which is determined based on the following: the width of the spray assembly, the characteristics of the imaging system used to scan the field, the distance separating the field portion observed by the imaging system from the spray system, and the forward speed of the spray assembly.

[0044] The method described in this paper can take into account the non-uniformity of the two-dimensional (2D) liquid spatial distribution of the nozzle. This 2D non-uniform liquid spatial distribution can be used when defining the pattern of a point spray.

[0045] The methods described herein improve the efficiency and adaptability of targeted spraying. By precisely controlling the delivery of agrochemicals to the target and avoidable objects, the total amount of agrochemicals sprayed can be reduced, thereby increasing efficiency and effectiveness and reducing unwanted side effects, such as toxicity to crops and the environmental impact of agrochemical use, manufacturing, and transportation.

[0046] Figure 1 A schematic diagram of a first example targeted spraying system 100 is shown. System 100 includes a targeted spray control system 102, a spray assembly 104, and an imaging system 106. The targeted spraying system 100 may also include a distance detection unit 108. It should be understood that the targeted spraying system 100 may include... Figure 1 Other components not shown, such as one or more sensors (e.g., accelerometer, GPS receiver, sensors, etc.).

[0047] The spray assembly 104 includes a plurality of nozzles 110 mounted on a spray bar 112. The nozzles 110 are mounted along the spray bar 112 at regular intervals s. Each nozzle 110 has an associated electromechanical valve 114 located between the nozzle 110 and the spray bar 112. These electromechanical valves 114 can be quickly and precisely switched on and off to control the fluid flow from each nozzle 110 and produce each point spray. A point spray refers to the fluid output from the nozzle 110 during a single valve opening (i.e., between the valve opening and subsequent re-closing). The time during which the valve 114 opens and produces a single point spray can be referred to as the point duration or open duration t, and can be in the range of 3 ms to 20 ms. The spray bar 112 is in fluid communication with a tank and pressure system 116 via an optional inlet electromechanical valve 118. The tank and pressure system 116 includes a tank for containing the agricultural chemicals to be sprayed and a pressure system for generating and controlling pressure p, at which the agricultural chemicals are supplied from the tank to the spray bar 112 and ultimately to the nozzle 110. The volume of fluid output in a single point spray is a function of the on-time, nozzle design (e.g., nozzle diameter), and pressure, and the area covered by a single point spray (and therefore the dose per unit area) also depends on the distance between the nozzle and the surface (e.g., the ground). An inlet valve 118 allows the tank and pressure system 116 to be isolated from the spray bar 112, for example, for maintenance purposes.

[0048] It should be understood that the spray assembly 104 may differ from... Figure 1 The spray assembly is shown. For example, valve 114 can be integrated into spray bar 112, there can be more than one spray bar 112 and / or inlet valve 118 can be omitted.

[0049] Imaging system 106 scans a portion of the field in front of spray boom 112, passing through the area, and identifies target objects (for spraying) and avoidable objects (that should not be sprayed). The portion of the field scanned at any given time can be referred to as the spray window, and its size depends on the width of the imaging system and the spray system, the distance between the portion of the field observed by the imaging system and the spray system, and the forward speed of spray assembly 104. Scans can be performed continuously or periodically (e.g., every 100 ms). In the case of periodic scanning, depending on the time between scan operations, the resulting spray windows can overlap (in the direction of forward movement of the spray assembly) or be continuous. Scan operations are performed frequently enough to avoid gaps between spray windows. As mentioned above, depending on the type of agrochemical being sprayed, the target object can be a desired plant (i.e., crop) or an undesirable plant (i.e., weed). Similarly, avoidable objects can be desired plants (i.e., crop) or undesirable plants (i.e., weed). The imaging system 106 may include one or more cameras and / or other sensors, as well as a processing system arranged to process the data captured by the cameras and / or other sensors and output data defining the target objects to be sprayed and the objects to be avoided.

[0050] In the included cases, distance detection unit 108 determines the distance between spray boom 112 and the target object (i.e., the highest part of the plant to be sprayed) for one or more points (e.g., each point on the ground). For many applications, agricultural chemicals are sprayed substantially vertically downwards from nozzle 110, and the distance determined by distance detection unit 108 is the vertical height. In other spraying orientations (e.g., spraying a substantially vertical surface), distance detection unit 108 still determines the distance between spray boom 112 and the target object, but this can be, for example, in a vertical plane. Any suitable technique can be used to perform distance detection (e.g., height detection), and in the example, distance detection unit 108 may include a 3D depth sensor or a distance range sensor. Distance detection unit 108 may determine a single distance over the entire spray boom (e.g., the minimum distance between the spray boom and the target object below the spray boom), or it may determine more granular distance data, such as per group of adjacent nozzles, per nozzle, or even more granular.

[0051] The targeted spray control system 102 generates control signals for the valve 114 associated with the nozzle 110 in the spray assembly 104 based on inputs received from the imaging system 106 and optionally from the distance detection unit 108. The targeted spray control system 102 performs the improved methods for controlling the targeted spraying of agricultural chemicals described herein, and these will be described in more detail below.

[0052] Although the distance detection unit 108, the imaging system 106, and the fixed-point spray control system 102 are in Figure 1 While shown as separate elements, it should be understood that some or all of them may share common components (e.g., distance detection unit 108 and imaging system 106 may share sensors, and any or all of them may share processing capabilities), or two or more of distance detection unit 108, imaging system 106, and point spray control system 102 may be combined.

[0053] Figure 2 A schematic diagram of a second example spot spraying system 200 is shown. (Compared to...) Figure 1 Similar to the system shown, system 200 includes a fixed-point spray control system, spray components, an imaging system, and an optional distance detection unit, although in Figure 2 Only some of these components are visible. Specifically, Figure 2 The nozzle 110 and spray bar 112, as well as the camera 202 as part of the imaging system, are shown. The tank and pressure system, the targeted spray control system, and other parts of the imaging system may be located within the main body 204 of the targeted spray system 200. In this example, the targeted spray system 200 is towed behind a vehicle 206 (e.g., a tractor), and the direction of travel of the vehicle is indicated by arrow 208 as it moves forward. Figure 2 As shown, the imaging system scans a portion of the field 210 before the spray boom 112 passes through the area, i.e., the sprayed area 212 is behind the scanned area 210. Since the distance between the scanned portion of the field 210 and the sprayed area 212 is fixed, the time delay between the scanning of the field and the spray boom passing through the area can be calculated if the forward speed of the vehicle 206 is known. This is taken into account when generating control signals for the electromechanical valve 114 to spatially synchronize the point spray with the target object and the avoidance object. The scanned portion of the field 210 is the spray window, and although in Figure 2 It is not visible in the center, but the width of the spray window corresponds to the width of the spray boom 112. To observe this width completely, several cameras 202 are usually required (although in...). Figure 2 Only one camera (202) is visible in the middle.

[0054] Figure 1 System 100 and Figure 2 The spray assembly shown in system 200 each includes a single spray bar 112, with nozzles 110 mounted along the spray bar 112 at regular intervals s. In some example systems, the spray assembly may include multiple spray bars, wherein nozzles on different spray bars are offset relative to each other. Figure 3In the example shown, each of spray bars 302 and 304 has nozzles installed at a regular spacing s, but the nozzles are offset between the spray bars, thus giving the effective nozzle spacing s' = s / 2 for the entire spray assembly. The effective nozzle spacing can be further reduced by using more than two spray bars, for example, s / 3 for three spray bars, s / 4 for four spray bars, and so on. The spray bars are oriented perpendicular to the system's direction of travel, so that when multiple spray bars are present in the spray assembly, these spray bars are... Figure 3 Arrow 308 indicates that the directions of travel are spaced apart from each other. The control method described herein can be used with any arrangement of spray booms and nozzles. Of course, when valve control signals are directed to several parallel booms, the control signals for the second or subsequent booms must be delayed accordingly with the boom's forward speed, so that the spray points of all booms are positioned on the same transverse line on the ground.

[0055] The static 2D liquid spatial distribution of the nozzle, where it is not assumed to be uniform, can be mathematically defined, for example, by defining parameters of one or more equations, such as a Gaussian or normal distribution corresponding to the distribution in 2D. The static 2D liquid spatial distribution of the nozzle can be defined using any form of mathematical equation. Figure 4 Two distinct examples of one-dimensional (1D) point-dose profiles 402 and 404 are shown. These 1D profiles 402 and 404 can represent along a path passing through, for example... Figure 5 The dose profile shown is intercepted along a line (e.g., along line X-X' or line Y-Y') of the elliptical, targeted spray 502. In other examples, the static 2D liquid spatial distribution of the nozzle input to this method can be entered in the form of a pre-calculated lookup table. For example, as... Figure 6 As shown in Example 602, a lookup table can subdivide a 2D spatial distribution into a 2D grid of cells and specify the dose received within each cell. In addition to showing example lookup table 602, Figure 6 It also shows two vertical lines passing through the center of the fixed-point spray (e.g., along...). Figure 5 The equivalent 1D distributions of the lines X-X' and Y-Y' shown are 604 and 606.

[0056] Figures 4-6 The non-uniform 2D liquid space distribution shown assumes the nozzle is static and depends on the nozzle design; however, when the spray assembly moves, the final 2D liquid space distribution of any stationary spray is a modified version of the static 2D liquid space distribution, resulting from movement during the time the valve associated with the nozzle is open. As mentioned above, the valve opening duration can range from 3ms to 20ms, and the forward velocity of the spray assembly can be approximately 2m / s, causing the nozzle to move approximately 1cm to 4cm during the valve opening time. Figure 7 Two examples of the resulting 2D liquid space distributions 702 and 704 of a fixed-point spray, as a result of different opening durations, are shown for a given nozzle advance velocity. In the first example of the resulting 2D liquid space distribution 702, the opening duration is 5 ms, while in the second example of the resulting 2D liquid space distribution 704, the opening duration is 15 ms. Arrow 706 indicates the direction of movement of the spray assembly, and in the second example, the increased dose and increased diffusion due to the increased opening duration are clearly visible.

[0057] When the spray assembly moves in a direction perpendicular to the spray bar (e.g., forward), each nozzle has the same speed; however, if the movement of the spray assembly includes some rotation (e.g., because it is turning through a corner), the speeds of the nozzles on the spray bar will differ. Therefore, even if the same static 2D liquid space distribution is used for each nozzle, the resulting 2D liquid space distribution may differ between the nozzles.

[0058] In some examples, the same static 2D liquid spatial distribution can be used regardless of the distance between the spray boom 112 and the target object (as determined by the distance detection unit 108). In other examples, the static 2D liquid spatial distribution used depends on the detected distance. For example, when calculating the combined dose, the same static 2D liquid spatial distribution can be modified based on height (e.g., in a manner similar to modifications performed based on the forward movement of the spray assembly). In other examples, different static 2D liquid spatial distributions can be stored and used for different distances between the spray nozzles and the target object. Increased distance results in a larger (i.e., a larger scale) point spray, but a lower dose per unit area because the volume of spray remains the same but is distributed over a wider area. In some examples, a single detection distance (e.g., a single distance between the spray boom and the target object) can exist for all nozzles, provided by the distance detection unit 108; however, in other examples, finer-grained distance data, such as distance data for each nozzle, can be provided. Therefore, even if the same static 2D liquid space distribution is used for each nozzle and even if all nozzles travel at the same speed, the resulting 2D liquid space distribution may differ between nozzles (e.g., in cases where the surface is uneven and / or the target object is at a different height).

[0059] Since the expansion of a static 2D liquid spatial distribution depends on the translational velocity (e.g., forward velocity) of the spray assembly, in some examples, the nozzle opening duration can be adjusted to compensate for velocity variations (e.g., velocity variations determined based on motion data input to the method). However, because changing the opening duration alters the dosage, even when using this method, individual spray points may have different maximum applied doses. This means that the combined dose of the point sprays and their arrangement will vary depending on the local nozzle velocity.

[0060] Figure 8 A point spray control system (such as) is shown. Figure 1 An example operation method of the fixed-point spray control system 102 shown. Figure 7 As shown, the method includes identifying both the target object and the avoidance object approaching the target object (box 802). This identification (in box 802) can be performed by the imaging system 106. The identified nearby avoidance object can be within a predefined range of the target object, for example, touching the target object or within a few centimeters of the target object. Using the detected position data of the target object and the avoidance object, the method then defines a pattern of targeted spray that meets two predefined criteria: (i) the maximum non-zero dose applied to the nearby avoidance object is not exceeded, and (ii) ideally, at least the minimum dose is applied to all parts of the target object (box 804). Control signals are then generated for the defined pattern (box 806), and these control signals are then output to the spray control system. The control signals may include control signals for each nozzle in the spray assembly and define when the nozzle opens and when the nozzle closes (which in turn defines the opening time, i.e., the time the nozzle is open, and the opening duration, i.e., the length of time the nozzle is open). These time values ​​further specify the position of the targeted spray (based on the forward velocity of the spray assembly) and the size of the targeted spray (because this depends on the opening duration, which is the interval between the opening and closing of the nozzle). These control signals enable the spray control system to deliver the defined fixed-point spray pattern.

[0061] The minimum dose defined for the target object and used in standard (ii) above is a local dose, i.e., the minimum dose to be applied to all parts of the target object. This minimum dose can be defined based on the amount of agrochemical per unit area (e.g., in g / cm²). 2 or ml / cm 2 (unit)

[0062] The maximum dose defined for adjacent avoidance targets and used in criterion (i) above can be a local dose (i.e., the maximum dose that should be applied to any part of the target target) or a global dose (i.e., the maximum dose that should be applied to the target target as a whole). For the local maximum dose, it can be based on the amount of agrochemical per unit area (e.g., in g / cm²). 2 or ml / cm 2 The maximum global dose can be defined based on the absolute amount of the agrochemical on the target plant (e.g., in g or cm³). 3 The absolute quantity is defined as the amount of agrochemical per unit area (e.g., in g / cm³). 2 or ml / cm 2 The area is calculated by multiplying the total area of ​​the plants by the area of ​​the plant (in units). The total area of ​​the plants can be determined by the imaging system 106 (e.g., using image analysis).

[0063] As mentioned above, the maximum dose is non-zero (i.e., it is greater than zero), which means that some agricultural chemicals will be sprayed onto the object to be avoided, and the maximum dose can be limited to avoid negative effects such as death / damage to the object to be avoided or unwanted (or unsafe) levels of residues in the plant (e.g., in the case of the object to be avoided being a crop plant) or growth promotion (e.g., in the case of the object to be avoided being a weed).

[0064] It should be understood that (in box 802) there may be more than one target object and / or avoidance object identified. In the case of more than one target object, a single pattern can be defined to satisfy the minimum dose criterion (ii) for all target objects. Alternatively, separate patterns can be defined, one pattern per target object. In the case of more than one avoidance object, one or more defined patterns collectively satisfy the maximum dose criterion (i) for each avoidance object. It should also be noted that this method is applicable when different types of target objects or avoidance objects exist, each type requiring a different level of minimum dose (for target objects) or maximum dose (for avoidance objects). The imaging and real-time object classification system can determine the type of object and thus identify the corresponding minimum or maximum dose for that object. For example, if several weeds are present in a crop field and spraying is required, a minimum dose to be applied can be defined for each weed. Another case is when several avoidance plants exist in the same field; a maximum dose can be defined for each species, for example, to take into account their sensitivity to the chemicals used. If one species is less susceptible to damage than another, it is preferable to adjust the tolerance level so that the minimum amount applied to the target plant in direct proximity can be used.

[0065] The same modifications to the minimum and maximum dosages can be applied based on the size of the object to be sprayed. For example, small crop plants may be much less resistant to herbicides than well-developed plants, so the maximum allowable dosage on such small plants would be defined as lower than the maximum allowable dosage on larger, avoidable plants.

[0066] To define the pattern (in box 802), the positions of one or more spot sprays are determined using the locations of the target and avoidance objects. The 2D liquid spatial distribution of the spot sprays is then used to calculate the dose applied to each object (e.g., by summing the contributions of spot sprays from different locations). These calculated doses are then compared to two criteria, and the positions of one or more spot sprays can be adjusted and / or the number of spot sprays can be changed (by adding or removing one or more spot sprays). As mentioned above, the 2D liquid spatial distribution of the spot sprays can be non-uniform, can depend on the advance speed and valve opening duration (i.e., the time between valve opening and subsequent closing), and can be mathematically defined or stored in a lookup table. Figure 9 , Figure 12A , Figure 12B , Figure 13A , Figure 13B and Figure 19 Various different methods for defining a pattern (in box 804) are shown in the figure and described below.

[0067] Figure 9 A first example method is shown for defining a spot spray pattern that satisfies the criteria detailed above (in box 804). In this first example method, a first spot spray is placed as close as possible to a neighboring object to avoid (box 902), without exceeding the maximum non-zero dose on the object to avoid. Then, one or more additional spot sprays are placed as close as possible to neighboring objects to avoid to form a continuous spray area with at least a minimum dose on a portion of the target object, while the combination of the first spot spray and these newly placed additional spot sprays does not exceed the maximum non-zero dose on the object to avoid (box 904). Finally, if and if necessary, one or more additional spot sprays are placed further away from neighboring objects to extend the continuous spray area so that it completely covers the target object (box 906). Within the continuous spray area, the dose at any point is not less than a predefined minimum dose. If more than one target object has been identified (in box 802), the method can be repeated until all target objects have been treated ("Yes" in box 908). The term "treated" is used in... Figure 9In the context of this, it refers to placing a point spray on a target object (i.e., performing boxes 902-906 against that target object). A continuous spray area on one target object may overlap with or not overlap with a continuous spray area on another target object, meaning that one or more spray areas are provided from all placed points (for all target objects, in boxes 902-906).

[0068] exist Figure 9 In the method shown, the position of the fixed-point spray is limited by the nozzle spacing on an axis perpendicular to the system's direction of travel (and therefore parallel to the spray bar); or in the case where the spray system contains multiple spray bars (e.g., as shown in the diagram). Figure 3 As shown), this is limited by the effective nozzle spacing. This axis, perpendicular to the system's direction of travel, can be referred to as the lateral axis, and the position of the pinpoint spray along this axis can be referred to as the lateral offset (e.g., the starting point relative to one end of the spray boom). The longitudinal offset is along a direction parallel to the system's direction of travel (and perpendicular to the spray boom), and each longitudinal offset corresponds to the duration for which the spray assembly moves at a known speed. This speed, corresponding to the speed on the surface where the target object is located, can be referred to as the displacement speed or forward speed, to distinguish it from the speed at which the nozzles are opened and closed by controlling the corresponding electromechanical valves. The lateral offset defines which nozzle on the spray boom is used (and therefore which electromechanical valve is turned on and off), while the longitudinal offset defines the time interval of the pinpoint spray and thus defines the control signal.

[0069] In addition to defining the location of each point spray, the nozzle's open duration is also defined. As mentioned above, the 2D liquid space distribution of the point spray depends on its open duration. In some examples, the same open duration can be used for all point sprays, while in other examples, the open duration can differ between point sprays.

[0070] You can refer to this. Figure 10 and Figure 11 Further description Figure 9 The method. Figure 10 The objects identified within the spray window are shown, and in the example shown, this includes three identified target objects 1001-1003 and an identified avoidable object 1004. Figure 9 Line 1005 is also shown, indicating the path of the nozzles; therefore, any point spray must be centered on one of these lines laterally. In the example shown, there are eight lines 1005, each corresponding to one of the nozzles on the spray bar. It should be understood that there can be a different number of nozzles on the spray bar, and in some examples, there may be more than eight nozzles on the spray bar.

[0071] Place the first targeted spray 1006, which overlaps with the first target object 1001 and is as close as possible to the avoidance object 1004, without exceeding the predefined maximum non-zero dose of the avoidance object (in box 902). Figure 10 In the diagram, the targeted spray is shown as two concentric ellipses, where the outer ellipse indicates the maximum range of the spray, and the inner ellipse indicates the area where the spray dose exceeds a predefined minimum dose for the target object. Since the minimum dose for the target object will always be higher than the maximum non-zero dose for the object to be avoided, the first targeted spray 1006 is positioned such that the outer ellipse overlaps with the object to be avoided, but the inner ellipse does not. For this first target object 1001, an additional targeted spray 1008 (in box 904) is then placed to extend the area of ​​the target object 1001 receiving at least the minimum dose, while not exceeding the maximum non-zero dose on the object to be avoided.

[0072] exist Figure 10 In the example shown, the initial placement of the two fixed-point sprays 1006, 1008 provides a continuous spray area covering at least a minimum dose of the first target object 1001. Then, as... Figure 11 The method is repeated for the next target object 1002 (No in box 908). To cover the second target object 1002, a first spot spray 1101 is placed near the object to be avoided (in box 902), then a second spot spray 1102 is also placed near the object to be avoided (in box 904), and then a third spot spray 1103 is placed away from the object to be avoided, so as to extend the continuous spray area onto the target object 1002 (in box 906).

[0073] exist Figure 9 In the example methods shown and described above, the maximum non-zero dose defined for the object to be avoided can be a local dose, i.e., the dose per unit area.

[0074] Figure 12A A second example method is shown that defines a targeted spray pattern that satisfies the criteria detailed above (in box 804). This second example method is... Figure 9 The variations shown and described above, and in cases where there may be more than one target object close to the same avoidance object (such as... Figure 10 and Figure 11 (as shown in the example) and is particularly useful when the maximum non-zero dose is the global maximum dose. This is especially useful when only one target object is approaching any avoidable object (No in box 1202). Figure 12A The method to be with Figure 9 It runs in the same way as the method.

[0075] like Figure 12AAs shown, if it is determined that there is more than one target object approaching any single identified avoidance object ("Yes" in box 1202), then the global maximum total dose for the avoidance object is determined (in box 1204). This can be determined, for example, by looking up (e.g., in a table of the specific agrochemicals used). The global maximum total dose can depend on the plant type of the avoidance object or other criteria. As mentioned above, the plant type can be determined by the imaging system 106 (e.g., using image analysis) and provided as input. Figure 12A The method involves determining a global maximum value for the object to be avoided, then dividing this global maximum value into allocations for each target object that is close to the object to be avoided (box 1206). This can be achieved by dividing the global maximum value by the number of neighboring target objects, ensuring that each target object receives the same allocation. Alternatively, different doses can be allocated to different neighboring target objects, for example, where the global maximum value can be allocated among neighboring target objects based on one or more criteria, such as: the distance between the target object and the object to be avoided (where a larger dose is allocated to the first target object that is closer to a neighboring object than the second target object), the size of the target object (where a larger dose is allocated to the first target object that has a larger area than the second target object), etc.

[0076] After the allocation is performed (in box 1206), the method is as described above. Figure 9 Continuing as described, the difference is that when processing a target object and placing a spot spray close to a neighboring object to avoid, the allocated portion of the global maximum total dose is used to determine the proximity at which the spot spray can be placed (in boxes 1212 and 1214).

[0077] Figure 12B It shows Figure 12A A variation of the example method shown and described above. Figure 12B The method considers both local maximum dose and global maximum dose, and then uses the more stringent one (i.e., the one corresponding to a smaller amount of agrochemical). For example... Figure 12B As shown, if the global maximum dose results in a higher dose per unit area than the local maximum dose ("Yes" in box 1205), then the method is as follows: Figure 9 As shown and as described above (i.e., using the maximum local dose). However, if the global maximum dose results in a local dose lower than the local maximum dose ("No" in box 1205), then the method proceeds as follows. Figure 12A As shown and as described above (i.e., by creating and using allocations for each target object).

[0078] Figure 13AA third example method is shown for defining a spot spray pattern that satisfies the criteria detailed above (in box 804). In this third example, instead of adding spot sprays to create a continuous spray area covering the target object, an array of spot sprays is defined that provides a continuous spray area over the entire (or substantially the entire) spray window, and then the spot sprays are removed from the array to satisfy the two criteria (i) and (ii) above.

[0079] like Figure 13A As shown, the method includes determining the spray size and stride such that when the point sprays are staggered, no area receives a dose less than a minimum dose (box 1302). The stride (also referred to as pitch) defines the offset of the point sprays in the lateral and longitudinal directions. As mentioned above, the offset in the lateral direction is constrained to a multiple of the nozzle spacing. The point size and / or offset can be predefined and stored in a lookup table, such that determination (in box 1302) involves performing a lookup. The lookup table can define combinations of spray size and offset for different spray bar heights (e.g., different distances between the nozzle and the target object), so the lookup can involve identifying the appropriate value corresponding to the current spray bar height from the lookup table. Different lookup tables may exist for different spray bar types, nozzle arrangements (e.g., in terms of nozzle spacing and / or nozzle type), and / or agrochemicals. In some examples, the spray size can be fixed, and only the stride (in box 1302) is determined. The spray size can correspond to a specific nozzle opening duration.

[0080] Figure 13A The method considers each identified target object within the spray window sequentially. Once the spray size and stride have been determined (in box 1302), a fixed-point spray array is positioned across the entire spray window with the determined size and stride (box 1304), and an example of this is shown in... Figure 14 As shown in the image. Figure 14 The diagram shows the lateral stride T, which in this example is twice the nozzle spacing (or twice the effective nozzle spacing), and the longitudinal stride L. It should be understood that the spray window (and therefore the array) can be larger than... Figure 14 The spray windows shown are much larger. For example, an array covering the spray windows can include dozens of rows of spot spray (e.g., 10-20 rows or even 50 rows, depending on the forward speed of the spray components).

[0081] Then, one or more spot sprays are removed from the array based on various criteria (boxes 1306-1310). First, all spot sprays in the array that do not reach target object 1502 (or the first target object if multiple target objects are identified within the spray window) are removed from the array (box 1306). Figure 15 The removal step is shown. Figure 14The effect of the array shown is illustrated, and it can be seen that in this example, only five point sprays 1504 are retained. Then, the point sprays are removed one by one until the dose received by the adjacent avoidance object 1506 is less than the maximum non-zero dose (box 1308). Figure 16 The removal step is shown. Figure 15 The effect of the remaining portion of the array is shown, and it can be seen that in this example, only three spot sprays 1504 are retained. In the presence of multiple adjacent avoidable objects, all of these objects are considered when removing spot sprays (in box 1308). Finally, any spot sprays that are unnecessary for achieving a minimum dose across the entire target object are removed (box 1310). Figure 17A The removal step is shown. Figure 16 The effect of the remaining portion of the array shown is illustrated, and it can be seen that in this example, only two point sprays 1504 are retained. The output of the three removal operations (boxes 1306-1310) for the target object is the array portion of that target object.

[0082] Figure 13A The method considers each identified target object within the spray window sequentially. Therefore, after all removal operations have been performed on the first target object (in boxes 1306-1310) and an array portion has been generated for the first target object, the method is repeated for each other identified target object ("No" in box 1312, followed by boxes 1304-1310) until all target objects have been processed ("Yes" in box 1312) and an array portion has been generated for each identified target object within the spray window. As mentioned above, to generate array portions for different target objects, each object starts with the same array (in box 1304), but different spot sprays can be removed (in boxes 1306-1310), so the array portions for different target objects will be different. Once array portions have been generated for all target objects ("Yes" in box 1312), these array portions are combined into a single pattern to produce the resulting pattern for all target objects (box 1313). When combining the array portions, the patterns are added together to produce the resulting pattern of spot spray, and any duplicate spot sprays are removed (i.e., so that only one spot spray exists at any particular location in the resulting pattern). Figure 17B The resulting patterns for all target objects are shown, and it can be seen that the resulting patterns include two fixed-point sprays for the first target object 1502, four fixed-point sprays for the second target object 1704, and four fixed-point sprays for the third target object 1706.

[0083] At this point in the method, it is determined whether all array start positions have been analyzed (box 1314). These start positions can be defined based on the longitudinal offset L' of the first row in the fixed-point array from the start point of the spray window, such as... Figure 18 As shown. Figure 18 Two distinct starting positions, L1' and L2', are shown, and there may be one, two, or more candidate starting positions (i.e., values ​​of L'). Before identifying and selecting the optimal solution, [the following is used]... Figure 13A The method is used to analyze it. If there are more candidate (i.e., possible) starting positions to analyze ("No" in box 1314), then repeat. Figure 13A The method is the same (boxes 1304-1313), but the initial arrangement of the spot spray array is slightly different, specifically specified by the particular value of L' being analyzed (in box 1304). The only difference in the iterations of the method for different starting positions is the arrangement of the initial array (in box 1304). The same criteria are then used to remove one or more spot sprays, and due to the slightly different arrangement of the initial array (in box 1304), different combinations of spot sprays may be removed, thus producing different resulting patterns of spot sprays for each starting position once all target objects have been processed (as shown in the output from box 1313).

[0084] After determining the resulting pattern for each possible starting position ("Yes" in box 1314), the optimal resulting pattern is identified (box 1316), and this is the pattern for generating the control signal (in...). Figure 8 (In box 806). The optimal result pattern is the result pattern that leads to the most effective use of the agrochemical (i.e., when compared with other result patterns from different starting positions). A variety of different criteria can be used to evaluate the effective use of agrochemicals, such as one or more of the following: the amount of agrochemical sprayed on the avoidance object and / or the total amount of agrochemical sprayed. In either case, a smaller amount of agrochemical is better because it corresponds to a degree of waste and / or overuse (considering that the minimum dose applied to the target object has already been guaranteed as a result of earlier methodological steps).

[0085] Although Figure 13A The resulting pattern for all target objects is formed (in box 1313), but in other examples, the array portion and starting position of each target object can be considered independently, such as... Figure 13B As shown. In this variant, the starting position is considered on a per-target-object basis, and the optimal array portion is identified for each target-object (box 1318).

[0086] It should be understood that, although Figure 13A and Figure 13B The diagram shows that block 1308 occurs before block 1310, but in the variant shown, the order of these two steps can be reversed.

[0087] Although Figure 13AThis shows all the resulting patterns generated before performing the evaluation to identify the best resulting pattern (in box 1316), but in Figure 19 In a variation of the method shown, an evaluation can be performed each time a new result pattern is generated for a new starting position (in box 1313). Figure 19 As shown, once a new result pattern is generated, it is determined whether this result pattern is superior to the previously stored best result pattern (box 1913). This evaluation (in box 1913) uses the same criteria as described above (refer to box 1316), and it can be understood that for the first starting position of the spray window to be considered, there will be no previously stored best result pattern, so the result of this evaluation will be "yes". If the newly generated result pattern is superior to the previously stored best result pattern ("yes" in box 1913), the newly generated result pattern is stored to replace the previously stored pattern (box 1914), but if the newly generated result pattern is not better ("no" in box 1913), it is not stored. The method then continues to determine whether all starting positions have been analyzed (in box 1314), and repeats the method if necessary. Once all array starting positions have been analyzed ("yes" in box 1314), the stored best result pattern is output (box 1916). Figure 19 The method only requires storing one result pattern, and the comparison is simplified to a comparison of two result patterns (the current best pattern and the new pattern), which results in a faster comparison.

[0088] exist Figure 19 In the variations shown and described above, instead of storing only a single result pattern, each result pattern can be stored, but the result pattern currently considered the best is marked as best. Then, in the comparison (in box 1913), the marked result pattern is used to determine whether to mark the newly created result pattern (if it is better) and remove the mark from the currently marked pattern, or whether not to mark the newly created result pattern (if it is not better than the currently marked pattern).

[0089] Figure 19 The variant shown can also be applied to Figure 13B The method allows for the generation of array portions for all starting positions of the target object, rather than generating array portions for all starting positions before identifying the optimal array portion. Instead, the evaluation is performed each time a new array portion is generated (at a new starting position).

[0090] exist Figure 13A and Figure 19 In this method, an array portion is generated for each target object, and then they are combined (in box 1313). In contrast, another example method is shown that defines a targeted spray pattern that satisfies the criteria detailed above (in box 804). Figure 20In this context, all target objects are considered simultaneously. This means... Figure 20 The method generates the resulting pattern for the starting position without first generating an array portion for each target object.

[0091] like Figure 20 As shown, this method is used in conjunction with Figure 13A and Figure 19 The process begins in the same manner (boxes 1302-1304), but when performing the removal steps (boxes 2006-2010), all target objects and all objects to be avoided are considered. For example... Figure 20 As shown, all spot sprays that did not reach any target object are removed (box 2006). Then, spot sprays are removed one by one until the dose received by each avoided object is less than the maximum non-zero dose (box 2008), and any spot sprays unnecessary for reaching the minimum dose on any target object are also removed (box 2010). After generating the resulting pattern for the starting position in this way, the method is then performed as described above with reference to Figure 13.

[0092] Figure 12A or Figure 12B The modifications shown and described above can also be applied. Figure 13A , Figure 13B and Figure 19 The method shown. Figure 19 The modifications shown and described above can also be applied. Figure 20 The method shown (where the input to box 1913 comes from box 2010).

[0093] Although Figure 13A , Figure 13B and Figure 20 Different starting positions for sequential analysis are shown, but it should be understood that in variations of the method shown, all starting positions can be analyzed in parallel to enable the identification of the optimal solution (in box 1316).

[0094] By using the methods described above, the efficiency and adaptability of targeted spraying are improved. By more carefully controlling the delivery of agrochemicals to the target organisms, the total amount of agrochemicals sprayed can be reduced, thereby increasing efficiency and effectiveness, and reducing the environmental impact of agrochemical use, manufacturing, and transportation.

[0095] The minimum dose used can be fixed or variable and can be provided as input to the method described above. In some examples, the targeted spray control system 102 can determine the minimum dose used by the method described above based on data received from the imaging system 106. Alternatively, this determination can be performed by the imaging system 106, and the result is input to the targeted spray control system 102. Figure 21As shown, target object data is received (box 2102). In the case where this method is executed by the point spray control system 102, this data is received from the imaging system 106. Based on the received data, a predefined minimum dose can be adjusted based on the size and / or type of the target object, or a predefined minimum dose can be selected from multiple candidate predefined minimum doses (box 2104). The resulting minimum dose is then output (box 2106) for use in the above method. The maximum non-zero dose can be provided as input to the above method, or used in conjunction with... Figure 21 The corresponding method is used to determine this.

[0096] For larger target objects, the minimum and / or maximum doses can be increased (in box 2104), where the size of the target object can refer to its area and / or its height (i.e., how tall it is). Additionally or alternatively, the minimum and / or maximum doses can depend on the type of plant being the target object.

[0097] By adjusting the minimum and / or maximum doses based on size and / or type, the dosage applied to the target and / or avoidable objects is customized for specific objects, and the overall effectiveness and / or efficiency of agrochemical application can be increased.

[0098] In the methods described above, many criteria involve ensuring that all parts of the target body receive at least a predefined minimum dose. In variations of the methods described above, the criteria can be modified such that at least a minimum proportion of the target body (e.g., at least 90% of the target body) receives the predefined minimum dose.

[0099] When the minimum dose standard is relaxed so that the resulting pattern does not guarantee that all target subjects receive at least the minimum dose, but rather that at least a predetermined minimum proportion of target subjects receive at least the minimum dose, the following applies: Figure 9 , Figure 12A , Figure 12B , Figure 13A , Figure 13B , Figure 19 and Figure 20 The resulting pattern generated by any of the methods in can be as follows: Figure 22 As shown, additional spot spraying is used to supplement it. (See example.) Figure 22 As shown, the portion of the target object that did not receive the minimum dose according to the resulting pattern is identified (box 2202), and then one or more additional targeted sprays are added to the resulting pattern so that all areas of the target object receive the minimum dose (box 2204). This generates an updated resulting pattern, which is then used to generate a control signal (in...). Figure 8 (in box 806).

[0100] (In box 2204) The added additional point sprays can have a different size than the point sprays in the input pattern; for example, the additional point sprays can be smaller. Alternatively, the point sprays can be positioned at lateral and / or transverse offsets, which differ from the offsets used to generate the point spray array (in box 1304). These additional point sprays do not form a regular pattern and can be used to fill in areas where... Figure 9 , Figure 12A , Figure 12B , Figure 13A , Figure 13B , Figure 19 and Figure 20 The "gap" in the resulting pattern generated by any of the methods.

[0101] The above method can be performed for each spray window, and as mentioned above, the spray windows can be continuous or partially overlapping. When the spray pattern is larger than the spray window in the forward direction, for example, when a large target object (such as a dense patch of weeds) must be sprayed, the pattern is placed at a fixed point selected in the first window (e.g., using...). Figure 8 The method will preferably continue in the second window, and the fixed point at the bottom of the first window will be used as a reference for placing fixed points that need to be added continuously in the second window. Referring back to the above method, this means that when analyzing different starting positions, if the target object extends from the first window to the second spray window, the starting position selected for the first spray window is automatically selected for the second spray window. Figure 13A , Figure 13B , Figure 19 and Figure 20 This means that for the second spray window, only a single starting position is analyzed, and this starting position is the same as the starting position used in the immediately preceding first spray window. Furthermore, when using... Figure 20 In the case of the method, for both the first and second spray windows, the same spray size and stride are used for the point spray array (as described in the positioning in box 1304).

[0102] Figure 24 Three partially overlapping spray windows 2401-2403 are shown. In the case of partially overlapping spray windows, as... Figure 24As shown, the above method can be performed based on a modified spray window, wherein the modified spray window excludes the portion of the spray window that overlaps with the next spray window. This means that the above method is initially performed for a first modified spray window 2411, which includes the first spray window 2401 but excludes the overlap 2404 with the second spray window 2402. By performing the above method in this way, a control signal (in block 806) is first generated and output for the first modified spray window 2411. The method is then repeated to generate a control signal for the next modified spray window 2412, which includes the second spray window 2402, which includes the overlap 2404 with the immediately preceding spray window (spray window 2401) but excludes the overlap 2406 with the next spray window (third spray window 2403). This operation is then repeated for all subsequent spray windows, wherein for each spray window, the portion overlapping with the previous window is considered, while the portion overlapping with the next window is excluded. As described above, when a target object extends from one modified spray window to the next, its placement in both modified spray windows follows the same pattern (e.g., using the same starting position, and using...). Figure 20 In the case of this method, the same spray size and stride are used for the array.

[0103] The above method can be implemented by a fixed-point spray control system, for example... Figure 1 The targeted spray control system 102 shown is implemented by one or more processors. These processors can be programmable (e.g., a central processing unit (CPU) or microcontroller), field-programmable gate array (FPGA), DSP, ASIC, PLC, and / or one or more ARM processors, etc. Figure 23 Various components of an example targeted spray control system employing a computation-based device 2300 are shown. As described above, this computation-based device can also perform... Figure 1 Some functions of the distance detection unit 108 and imaging system 106 shown.

[0104] The computing-based device 2300 includes one or more processors 2302, which may be a microprocessor, a controller, or any other suitable type of processor, for processing computer-executable instructions to control the operation of the device in order to perform the methods described herein (e.g., such as...). Figure 8 , Figure 9 , Figure 12A , Figure 12B , Figure 13A , Figure 13B and Figures 19-22(As shown). In some examples, such as some examples of systems using an on-chip architecture, processor 2302 may include one or more fixed functional blocks (also referred to as accelerators) that implement part of a method for controlling a targeted spray system in hardware (rather than software or firmware). Platform software, including operating system 1804 or any other suitable platform software, may be provided at the computing-based device to enable application software 2306 (such as software implementing the methods described herein) to execute on the device.

[0105] Computer-executable instructions may be provided using any computer-readable medium accessible to the computing device 2300. Computer-readable media may include, for example, computer storage media (such as memory 2308) and communication media. Computer storage media, such as memory 2308, include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other memory technologies, CD-ROM, Digital Universal Disc (DVD) or other optical storage, cassette tape, magnetic tape, disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. Conversely, communication media may embody computer-readable instructions, data structures, program modules, or other data in modulated data signals (such as carrier waves) or other transmission mechanisms. As defined herein, computer storage media does not include communication media. Although the computer storage medium (memory 1808) is shown within the computing-based device 2300, it should be understood that the storage can be remotely distributed or located and accessed via a network or other communication link (e.g., using communication interface 2310).

[0106] Communication interface 2310 may be configured to receive data used in the methods described herein, such as motion data (e.g., from...). Figure 2 The data includes the vehicle 206 shown (or a sensor within the point spray system), target object data (e.g., from imaging system 106), and altitude data (e.g., from distance detection unit 108). The communication interface 2310 may also be arranged to output the generated control signals (e.g., output to electromechanical valve 114 in spray assembly 104).

[0107] The computing-based device 2300 may also include an input / output interface 2312 arranged to output display information to a display device 2314 that may be separate from or integrated with the computing-based device 1800. For example, the display device 2314 may be attached to Figure 2The main body 204 of the targeted spraying system 200 shown is located in a vehicle 206. Alternatively, display information can be output via communication interface 2310 to a remote display device 1814 (at a monitoring location remote from the targeted spraying system). The display information can provide a graphical user interface. Input / output interface 2312 can also be arranged to receive and process input from one or more devices, such as user input device 1816 (e.g., ...). Figure 2 (One or more buttons on the body 204 of the targeted spraying system 200 shown, or located in the vehicle 206). This user input can be used to adjust parameters of the method or to provide input, such as predefined minimum and / or maximum doses. In embodiments, if the display device 2314 is a touch-sensitive display device, it can also serve as a user input device 2316. In some examples, instead of or in addition to the communication interface 2310, the input / output interface 2312 can be arranged to output generated control signals (e.g., output to the electromechanical valve 114 in the spray assembly 104).

[0108] The memory 2308 may be arranged to store data used by the methods described herein, such as static 2D liquid space distribution 2318 and configuration data 2320 of the spray assembly (e.g., nozzle spacing). The memory 2308 may be arranged to store the lookup table described above and the resulting pattern generated using the methods described above.

[0109] Although Figure 23 It shows that it can be used Figure 1 and Figure 2 The illustrated point-spraying systems 100 and 200 are implemented locally on a single computing device. However, in other examples, some processing and / or data storage may be implemented remotely to the spray assembly, for example, on a remote computing device that may be located in a data center or elsewhere. For example, static 2D liquid space distribution 2318 and / or configuration data 2320 of the spray assembly may be remotely stored and accessed via communication interface 2310. Additionally or alternatively, the determination of the resulting pattern may be performed via a remote computing device and control signals received by the spray system via communication interface 2310. In other examples, processing and / or data storage may be divided in different ways between a local computing device near the spray assembly and a remote computing device (or multiple local computing devices, e.g., where data processing is performed on a different computing device than the data storage).

[0110] The term "computer" is used herein to refer to any device that has processing power that enables it to execute instructions. Those skilled in the art will recognize that this processing power is incorporated into many different devices, and therefore the term "computer" includes PCs, servers, mobile phones, personal digital assistants, and many other devices.

[0111] Those skilled in the art will recognize that storage devices for storing program instructions can be distributed across a network. For example, a remote computer can store examples of processes described as software. A local or terminal computer can access the remote computer and download part or all of the software to run the program. Alternatively, a local computer can download fragments of software on demand, or execute some software instructions at a local terminal while executing others at a remote computer (or computer network). Those skilled in the art will also recognize that, by utilizing conventional techniques known to those skilled in the art, all or part of the software instructions can be executed by dedicated circuitry (e.g., DSPs, programmable logic arrays, etc.).

[0112] As will be apparent to those skilled in the art, any range or device values ​​given herein can be extended or altered without losing the desired effect.

[0113] It should be understood that the benefits and advantages described above may relate to one embodiment or several embodiments. The embodiments are not limited to those that solve any or all of the stated problems or those that have any or all of the stated benefits and advantages.

[0114] Any reference to an item “an” refers to one or more of these items. The term “comprising” is used herein to mean including the identified method block or element, but such block or element does not include a unique list, and a method or apparatus may include additional blocks or elements.

[0115] The steps of the methods described herein can be performed in any suitable order or simultaneously when appropriate. Furthermore, individual blocks can be removed from any method without departing from the spirit and scope of the subject matter described herein. Aspects of any example described above can be combined with aspects of any of the other examples described to form additional examples without losing the desired effect.

[0116] It will be understood that the above description of the preferred embodiments is given by way of example only, and various modifications can be made by those skilled in the art. Although various embodiments have been described above with some level of detail or with reference to one or more individual embodiments, those skilled in the art can make many changes to the disclosed embodiments without departing from the spirit or scope of the invention.

Claims

1. A method of operating a targeted spray control system for spraying agricultural chemicals onto a target object using a spray assembly and minimizing spraying onto avoidable objects, said spray assembly comprising an array of nozzles, wherein, The opening time and duration of the nozzle can be controlled individually, and the method includes: (i) Identify the target object and nearby avoidance objects (802); (ii) The pattern of the point spray is defined by the placement and opening duration of the point spray, such that the maximum non-zero dose applied to the adjacent avoidance object is not exceeded, and at least a minimum dose per unit area is applied to the target object at a predefined proportion (804), wherein each point spray is defined by a non-uniform two-dimensional liquid space distribution, the non-uniform two-dimensional liquid space distribution being modified by the distance from the nozzle to the target object, the forward speed of the spray assembly, and the opening duration of the nozzle; (iii) Generate a control signal (806) for the array of nozzles based on the pattern of the targeted spray; and (vi) Output the control signal to the spray assembly.

2. The method according to claim 1, wherein, Defining the pattern of the targeted spray includes, for each target object: Without exceeding the maximum non-zero dose on the adjacent avoidable object, the first spot spray is placed as close as possible to the adjacent avoidable object (902). One or more additional spot sprays are placed as close as possible to the adjacent avoidance object to form a continuous spray area of ​​at least the minimum dose on a portion of the target object, while not exceeding the maximum non-zero dose (904) on the adjacent avoidance object. and If the portion of the target object covered by the continuous spray area is smaller than the predefined proportion of the target object, one or more additional fixed-point sprays are placed such that at least the minimum dose of the continuous spray area extends over the predefined proportion of the target object (906).

3. The method according to claim 1, wherein, The pattern of the targeted spray is defined as follows: Determine the spray size and stride such that when the point sprays are staggered, no area receives a dose less than the minimum dose (1302). Multiple result patterns are generated, each corresponding to a different candidate starting position among multiple candidate starting positions of the fixed-point spray array; and Select the result pattern (1316) that minimizes waste of agricultural chemicals from the multiple result patterns. Each resulting pattern is generated in the following way: - For each target object, the array portion is generated as follows: -- Starting from the candidate starting position, position the fixed-point spray array (1304) across the entire spray window with the determined size and stride. -- Remove all spot sprays (1306) from the array that do not reach the target object; -- Remove the targeted spray until the dose received by the adjacent avoidance object is less than the maximum non-zero dose (1308); and -- Remove any spot sprays (1310) that are unnecessary for achieving the minimum dose at the predefined ratio of the target object. and - Combine the array portion (1313) of each target object.

4. The method according to claim 1, wherein, The pattern of the targeted spray is defined as follows: Determine the spray size and stride such that when the point sprays are staggered, no area receives a dose less than the minimum dose (1302); and For each target object, multiple array segments are generated, each array segment corresponding to a different candidate starting position among multiple candidate starting positions of the fixed-point spray array; and Select from the plurality of array sections the array section that minimizes waste of agrochemicals (1318). Each array section is generated in the following way: Starting from the candidate starting position, a fixed-point spray array (1304) is positioned across the entire spray window with the determined size and stride. Remove all targeted sprays (1306) from the array that did not reach the target object; Remove the targeted spray until the dose received by the adjacent avoidance object is less than the maximum non-zero dose (1308); and Remove any spot sprays (1310) that are unnecessary for achieving the minimum dose at the predefined ratio for the target object.

5. The method according to any one of the preceding claims, wherein, The maximum non-zero dose is the maximum non-zero dose per unit area.

6. The method according to any one of claims 2-4, further comprising: Determine if there is more than one target object approaching the avoidance object (1202). and In response to determining that there is more than one target object approaching the avoidance object, the maximum non-zero total dose for the avoidance object is divided into an allocation amount (1206) for each target object, wherein the allocation amount is the maximum non-zero dose used when defining the pattern for the target object.

7. The method according to any one of claims 2-4, further comprising: Determine if there is more than one target object approaching the avoidance object (1202). In response to determining that there is more than one target object approaching the avoidance object, determine whether the maximum non-zero total dose for the avoidance object is given a higher dose per unit area than the local maximum non-zero dose per unit area for the avoidance object (1205). and In response to determining that the maximum non-zero total dose for the object to be avoided does not give a dose per unit area higher than the local maximum non-zero dose per unit area for the object to be avoided, the maximum non-zero total dose for the object to be avoided is divided into an allocation amount (1206) for each target object, wherein the allocation amount is the maximum non-zero dose used when defining the pattern for the target object.

8. The method according to claim 1, wherein, The pattern of the targeted spray is defined as follows: Determine the spray size and stride such that when the point sprays are staggered, no area receives a dose less than the minimum dose (1302). Multiple result patterns are generated, each corresponding to a different candidate starting position among multiple candidate starting positions of the fixed-point spray array; and Select the result pattern (1316) that minimizes waste of agricultural chemicals from the multiple result patterns. Each resulting pattern is generated in the following way: Starting from the candidate starting position, a fixed-point spray array (1304) is positioned across the entire spray window with the determined size and stride. Remove all spot sprays from the array that did not reach any target object (2006); Remove the targeted spray until the dose received by each neighboring object to be avoided is less than the maximum non-zero dose (2008); and Remove any spot sprays that are unnecessary for achieving the minimum dose at a predefined proportion on any target object (2010). and Combine the array portions (1313) of each target object.

9. The method according to any one of the preceding claims, wherein, The predefined proportion of the target object is less than the entirety of the target object, and the method further includes: Identify the portion of the target object that did not receive the minimum dose (2202); and The pattern of the spot spray is updated by adding one or more additional spot sprays to the portion of the target object (2204).

10. The method according to any one of claims 1 to 8, wherein, The predefined proportion of the target object is the entirety of the target object.

11. The method according to any one of the preceding claims, wherein, The two-dimensional liquid space distribution of the fixed-point spray is determined based on the static two-dimensional liquid space distribution of the nozzle and the received motion data of the spray assembly.

12. The method according to any one of the preceding claims, wherein, The two-dimensional liquid spatial distribution of the fixed-point spray is determined based on the static two-dimensional liquid spatial distribution of the nozzle and the detected distance between the nozzle and the target object.

13. The method according to any one of the preceding claims, wherein, The two-dimensional liquid space distribution of the fixed-point spray is determined based on the static two-dimensional liquid space distribution of the nozzle and the opening duration of the nozzle.

14. The method according to any one of claims 11-13, wherein, The two-dimensional liquid spatial distribution of the targeted spray is defined in a lookup table.

15. The method according to any one of claims 11-13, wherein, The two-dimensional liquid spatial distribution of the targeted spray is defined using a mathematically defined distribution.

16. The method according to any one of the preceding claims further comprises: In response to changes in the nozzle's forward velocity during spraying, the nozzle's opening duration is adjusted to maintain a constant two-dimensional liquid spatial distribution for point spraying.

17. The method according to any one of the preceding claims further includes adjusting a predefined minimum dose based on the size or type of the target object (2104).

18. A computer program comprising instructions that, when executed by a computer, cause the computer to perform the method according to any one of the preceding claims.

19. A computer-readable medium having a computer program stored thereon according to claim 18.

20. A targeted spraying control system for spraying agricultural chemicals using a spray assembly, said spray assembly comprising an array of nozzles, wherein, The opening time and duration of the nozzle can be controlled individually, and the targeted spray control system includes: Processor (1802); One or more interfaces (1810, 1812), said one or more interfaces being configured to receive target object data and output control signals to the spray assembly; and Memory (1808), which is arranged to store a computer program that, when executed by the processor, enables the control system to: (i) Identify the target object and nearby avoidance objects (802); (ii) The pattern of the point spray is defined by the placement and opening duration of the point spray, such that the maximum non-zero dose applied to the adjacent avoidance object is not exceeded, and at least a minimum dose per unit area is applied to the target object at a predefined proportion (804), wherein each point spray is defined by a non-uniform two-dimensional liquid space distribution, the non-uniform two-dimensional liquid space distribution being modified by the distance from the nozzle to the target object, the forward speed of the spray assembly, and the opening duration of the nozzle; (iii) Generate a control signal (806) for the array of nozzles based on the pattern of the targeted spray; and (vi) Output the control signal to the spray assembly.

Images