Unmanned aerial vehicle

Abstract

The present invention relates to an unmanned aerial vehicle (UAV) for agricultural weed management. The UAV includes a control processing unit (20) and a camera (30). The control processing unit is configured to control the UAV to fly within the height of a crop but below the vertical height of the crop and / or between rows of the crops but below the vertical height of the crops. The control processing unit is configured to control the camera to acquire at least one image of the ground within the height of the crop but below the vertical height of the crop and / or between rows of the crops but below the vertical height of the crops. The control processing unit is configured to analyze the at least one image to determine the presence and location of at least one weed on the ground.

Description

[Technical Field]

[0001] The present invention relates to unmanned aerial vehicles (UAVs), base stations, systems and methods for agricultural weed management, and computer program products for agricultural weed management. [Background technology]

[0002] The general background of the present invention is the assessment and management of agricultural land from a weed perspective. Currently, remote sensing and unmanned aerial vehicles (UAVs) such as drones do not acquire images at the resolution and quality required to perform the required imaging diagnostics.

[0003] WO 2019 / 076758 A1 proposes a UAV configured to hover in a stationary position above the crop, with a camera that can be lowered toward the crop and even into the crop stature to capture images. A drawback of this solution is that the downdraft from the propellers in this UAV configuration significantly moves leaves within the crop stature at the image capture point. Furthermore, the camera can easily become tangled within the crowded stature. Additionally, raising and lowering the camera from the UAV slows the field reconnaissance process, making it very time-consuming to acquire the necessary image data. Summary of the Invention

[0004] It would be advantageous to have improved means for agricultural weed management.

[0005] The object of the present invention is solved by the subject matter of the independent claims, further embodiments are incorporated in the dependent claims. It is to be noted that the aspects and examples described below with respect to the present invention also apply to unmanned aerial vehicles for agricultural weed management, to base stations, to systems and methods for agricultural weed management, as well as to computer program products.

[0006] According to a first aspect, an unmanned aerial vehicle (UAV) for agricultural weed management is provided. The UAV includes a control processing unit and a camera. The control processing unit is configured to control the UAV to fly to a position within the height of a crop but below the vertical height of the crop, and / or to a position between rows of the crop but below the vertical height of the crop. The control processing unit is configured to control the camera to acquire at least one image of the ground within the height of the crop but below the vertical height of the crop, and / or between rows of the crop but below the vertical height of the crop. The control processing unit is configured to analyze the at least one image to determine the presence and location of at least one weed on the ground.

[0007] In other words, the UAV flies close to the crop to enable close-up sensing of the weeds. In this way, the UAV can create a detailed and accurate weed map of the crop field.

[0008] The UAV can acquire data by randomly scanning around the field. The UAV can also determine where to position itself to acquire data based on image processing, or the user can direct the positioning. Because the UAV can fly through the crop height, acquiring more information in a shorter time than other technical solutions such as those proposed in WO 2019 / 076758 A1, the UAV (or multiple similar UAVs) can acquire weed images and economically create weed maps across the entire field.

[0009] In one example, the at least one image includes a plurality of images, and the control processing unit is configured to control the camera to capture the plurality of images at positions corresponding to a plurality of different positions relative to the ground within the height of the crop but below the vertical height of the crop and / or between rows of the crop but below the vertical height of the crop.

[0010] In other words, the granularity of the weed map is improved by capturing images at different heights and positions within the crop height and / or at different heights and positions between rows of multiple crops.

[0011] In one example, the control processing unit is configured to analyze at least one image to determine a weed map that classifies the presence of weeds on the ground based on the density and spatial distribution of the weeds.

[0012] In other words, the images are used to analyze the frequency and zonation of weeds on a crop field, helping to identify whether the weeds are uniformly distributed, present in patches, or whether single weed plants are randomly distributed across the field.

[0013] In one example, the size of the UAV is approximately 1 cm to 20 cm, preferably approximately 2 cm to 10 cm, and more preferably approximately 4 cm to 8 cm, and the weight of the UAV is less than 200 g, preferably less than 100 g, and more preferably less than 50 g.

[0014] Thus, UAVs have a small size for optimal freedom of movement within the crop canopy and between crop rows below the vertical height of the crop. When flying within the canopy, such UAVs are less susceptible to wind disturbances because the crop at least partially shields the UAV from such impacts. Additionally, the small size and weight of UAVs provides greater safety, for example, in the event of an unwanted accidental collision.

[0015] In one example, the UAV includes a propeller guard configured to at least partially surround a propeller of the UAV, and preferably configured to completely surround the propeller.

[0016] In this way, the propeller is protected from, for example, flapping leaves, which could interfere with the rotational movement of the propeller.

[0017] In one example, the UAV includes at least one weed control unit configured to be activated at a location determined by the control processing unit based on image analysis regarding the presence and location of at least one weed.

[0018] Thus, the UAV can acquire images, analyze them, and control weeds, for example by spraying an appropriate herbicide, all in one operational sequence. This is more efficient than solutions such as those proposed in WO 2019 / 076758 A1. In addition, a UAV equipped with at least one weed control unit is closer to the target weeds, which provides greater accuracy in spraying the herbicide on the target. A second advantage is that ground effect reduces the energy required for flight and increases the UAV's range.

[0019] In one example, the UAV includes a position determining means configured to provide the control processing unit with at least one position associated with the camera when the at least one image associated with the ground was acquired.

[0020] In one example, the UAV includes a transceiver. The control processing unit is configured to utilize the transceiver to transmit information regarding at least one image associated with the ground and / or information regarding image analysis of the at least one image associated with the ground, as well as at least one location associated with the camera when the at least one image associated with the ground was acquired. The control processing unit is configured to receive instructions to fly the at least one weed to the determined at least one location and to receive instructions to activate the at least one weed control unit at the at least one location. The instructions are based on the at least one image associated with the ground and / or image analysis of the at least one image associated with the ground transmitted by the transceiver.

[0021] In other words, the UAV transmits information about the captured image and the location where it was captured to another processing unit, such as one located in a base station, such as a transport vehicle. The base station, which has a larger size and processing power, can receive image information from multiple UAVs in the field. The base station coordinates the flight paths for the multiple UAVs and transmits instructions for flying the multiple UAVs to locations where weed control is required. The optimal flight path is calculated by considering various aspects of the UAVs, such as the energy level of the UAV, the type of control unit, the remaining ability of the control unit to control a particular weed problem, the distance between the UAV's location and the weed confirmation location, etc. In this way, when calculating the appropriate flight path, the control processing unit at the base station can take into account the "specialization" of different UAVs on the crop field (e.g., one UAV fully specialized in scouting weeds on the ground, another UAV fully specialized in weed control, etc.).

[0022] According to a second aspect, a base station for multiple UAVs is provided. The base station includes a UAV parking unit including at least one UAV, a control processing unit, and at least one transceiver. The at least one UAV includes the control processing unit, a camera, and the transceiver. The UAV parking unit of the base station is configured to transport the multiple UAVs. The control processing unit of the UAV is configured to control the UAV to fly from the base station to a position within the height of a crop but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. The control processing unit of the UAV is configured to control the camera to capture at least one image related to the ground within the height of a crop but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. The control processing unit of the UAV is configured to transmit information relating to at least one image related to the ground to a base station using the transceiver. The control processing unit of the base station is configured to receive information relating to the at least one image related to the ground from the at least one UAV using the transceiver. The control processing unit of the base station is configured to determine the presence and location of at least one weed on the ground by analyzing the at least one image received from the at least one UAV.

[0023] Thus, due to the UAV's small size and limited capacity to process the acquired image information, the UAV can transmit the image information via wireless communication to a base station with greater processing power. The base station is capable of receiving the image reconnaissance data from the small UAV and performing detailed image analysis, such as by using artificial intelligence algorithms to identify weed problems in crop fields.

[0024] In one example, a base station for multiple UAVs includes at least one UAV. The at least one UAV further includes a position determining means. The position determining means is configured to provide the control processing unit with at least one position associated with the camera when the at least one image associated with the ground was acquired. The control processing unit of the UAV is configured to transmit, using the transceiver, information regarding the at least one position associated with the camera when the at least one image associated with the ground was acquired. The control processing unit of the base station is configured to receive, using the transceiver, information regarding the at least one position associated with the camera when the at least one image associated with the ground was acquired. The control processing unit of the base station is configured to determine the presence of at least one weed on the ground at the at least one position by analyzing the at least one image received from the at least one UAV.

[0025] In one example, a base station for a plurality of UAVs includes at least one UAV, the at least one UAV further including at least one weed control unit, and a control processing unit of the base station is further configured to determine instructions for flying the at least one UAV to at least one location where the presence of at least one weed has been determined, and to determine instructions for activating the at least one weed control unit at the at least one location.

[0026] In one example, a base station for multiple UAVs includes a UAV parking unit with the multiple UAVs. A control processing unit of the base station is configured to determine a weed map by analyzing at least one image received from at least one UAV. The weed map is a classification of weed presence on the ground based on the density and spatial distribution of the weeds. The control processing unit of the base station is further configured to determine a flight path map for the multiple UAVs as a function of the weed map and as a function of the weed classification based on the density and spatial distribution of the weeds on the ground. The control processing unit of the base station is further configured to determine instructions for flying the multiple UAVs to at least one location where the presence of at least one weed has been determined, and further to determine instructions for activating at least one weed control unit at the at least one location based on the determined flight path map.

[0027] The control processing unit of the base station then coordinates the activity of the multiple UAVs. Depending on the weeds present on the field and the corresponding weed map, the multiple UAVs can be directed to control weeds in different ways. For example, spraying can be performed in patches, parallel strips, or a target-to-target approach. A high number of weeds and a uniform distribution favors a strip spraying strategy, while a low number of weeds favors a target-to-target spraying strategy. Additionally, the presence of weed patches allows for a patch-based strategy in which only specific areas of the field are treated by the UAV's weed control units. The control processing unit of the base station determines the optimal strategy and determines a flight plan for the multiple UAVs to optimally treat the field.

[0028] In one example, a base station for multiple UAVs includes a control processing unit configured to control the multiple UAVs to operate as alternating groups, i.e., one group is in flight while another group is on a UAV parking unit.

[0029] In other words, while one group of UAVs is flying over crop fields to control weeds, another group is on a base station, preferably being serviced by a service unit, such as recharging batteries and / or refilling spray tanks.

[0030] According to a third aspect, there is provided a system for agricultural weed management, the system including a base station for a plurality of UAVs, the base station including a UAV parking unit including at least one UAV, a control processing unit, and and at least one transceiver. The at least one UAV includes a control processing unit, a camera, and a transceiver. The UAV parking unit is configured to transport multiple UAVs. The control processing unit of the UAV is configured to control the UAV to fly from the base station to a position within the height of the crops but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. The control processing unit of the UAV is configured to control the camera to capture at least one image related to the ground within the height of the crops but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. The control processing unit of the UAV is configured to transmit information related to the at least one image related to the ground to the base station using the transceiver. The control processing unit of the base station is configured to receive information related to the at least one image related to the ground from the at least one UAV using the transceiver. The control processing unit of the base station is configured to determine the presence and location of at least one weed on the ground by analyzing at least one image received from the at least one UAV.

[0031] According to a fourth aspect, there is provided a method for agricultural weed management, the method comprising: a) flying the UAV into a position within the height of the crop and below the vertical height of the crop, and / or between rows of multiple crops and below the vertical height of the multiple crops; b) acquiring, by a camera, at least one image related to the ground within the height of the crop and below the vertical height of the crop and / or between rows of the crop and below the vertical height of the crop; c) determining at least one weed and its location on the ground by analyzing the at least one image.

[0032] In a fifth aspect, there is provided a computer program product for controlling a UAV according to the first aspect, the computer program product being configured, when executed by a processor, to perform a method according to the fourth aspect.

[0033] Advantageously, any advantage provided by any of the above aspects applies equally to all other aspects, and vice versa.

[0034] The above aspects and examples will be apparent from and elucidated with reference to the embodiments described hereinafter.

[0035] Exemplary embodiments are described below with reference to the following drawings: [Brief explanation of the drawings]

[0036] [Figure 1] 1 shows a schematic configuration of an example UAV for agricultural weed management. [Figure 2] 1 illustrates a schematic configuration of an example base station for multiple UAVs. [Figure 3] 1 shows a schematic example of a system for agricultural weed management. [Figure 4] 1 shows a schematic example of a method for agricultural weed management. [Figure 5] 1 shows a schematic illustration of a detailed example of a base station (UAV delivery vehicle) and multiple UAVs on agricultural land. [Figure 6a] 1 shows a schematic example of multiple UAVs spraying herbicides on a crop field based on different weed maps. [Figure 6b]1 shows a schematic example of multiple UAVs spraying herbicides on a crop field based on different weed maps. [Figure 6c] 1 shows a schematic example of multiple UAVs spraying herbicides on a crop field based on different weed maps. DETAILED DESCRIPTION OF THE INVENTION

[0037] FIG. 1 illustrates an example of an unmanned aerial vehicle (UAV) 10 for agricultural weed management. The UAV includes a control processing unit 20 and a camera 30. The control processing unit is configured to control the UAV to fly within the height of a crop but below the vertical height of the crop and / or between rows of the crops but below the vertical height of the crops. The control processing unit is configured to control the camera to acquire at least one image of the ground within the height of the crop but below the vertical height of the crop and / or between rows of the crops but below the vertical height of the crops. The control processing unit is configured to analyze the at least one image to determine the presence and location of at least one weed on the ground.

[0038] In one example, the term "agricultural weed management" refers to field scouting activities to identify weeds on the ground in a field. If at least one weed control unit 50 is present, the term also includes management activities of the identified weeds.

[0039] In one example, a location within the height of a crop plant and below the vertical height of the crop plant refers to a location within and / or adjacent to the spatial configuration (three-dimensional geometry) of the above-ground parts of the crop plant, and not a location above the height of the crop plant.

[0040] In one example, a location between rows of crops and below the vertical height of the crops refers to a location between crop statures within different rows, but not above the crop plants. The crop statures may partially overlap at their tops, and the UAV may fly directly underneath them.

[0041] In one example, the camera includes a conventional image sensor operable to capture images of the ground at a resolution that allows for image processing to determine at least one weed.

[0042] In one example, the camera is a 360-degree panoramic camera, or a camera capable of capturing images in substantially 360 degrees.

[0043] In one example, the camera is configured to operate across the visible wavelength range. In one example, the camera is configured to operate in the near-infrared region. In one example, the camera is monochromatic. In one example, the camera is configured to capture color information, such as RGB. In one example, the camera is configured to capture hyperspectral information. In this manner, image analysis for automatically detecting at least one weed can be improved.

[0044] In one example, the camera further includes a light source, which may be multispectral, to aid in image acquisition.

[0045] In one example, the control processing unit is configured to determine the type of at least one weed by analyzing the at least one image.

[0046] In one example, the image analysis of the at least one image includes utilizing a machine learning algorithm that is applied to process the image to determine the at least one weed.

[0047] In one example, the machine learning algorithm includes a decision tree algorithm.

[0048] In one example, the machine learning algorithm includes an artificial neural network.

[0049] In one example, the machine learning algorithm has been trained based on a plurality of images, hi one example, the machine learning algorithm has been trained based on a plurality of images that include images of at least one type of weed on the ground.

[0050] In one example, a machine learning algorithm has been trained based on a plurality of images containing such images.

[0051] Thus, the control processing unit can execute image processing software, including a machine learning analyzer. For example, images of specific weeds on the ground are captured along with information related to the size of the weeds. The information is related to the geographic location in the world where the weeds are found and the time of year when the weeds are found. The weed names can also be tagged with the images of the weeds. A machine learning analyzer, which may be based on an artificial neural network or a decision tree analyzer, is then trained on the captured images. In this way, when a new image of the ground is presented to the analyzer, which may have a timestamp related to the time of year and a tagged geographic location, the analyzer determines the specific type of weed in the image by comparing the image of the weed found in the new image with images of different weeds it has been trained on, taking into account the size of the weeds and the location and time of year in which the weeds are growing. Thus, the specific location and size of the weeds on the ground can be identified. The control processing unit can access a database containing various types of weeds. This database is compiled from experimentally determined data.

[0052] In one example, image analysis involves identifying bird or small mammal nesting sites and their locations (for their future conservation).

[0053] In one example, the term "crop" refers to orchard crop plants and / or arable crop plants, preferably arable crop plants, where the orchard crop is selected from the group consisting of apples, pears (European and Asian), grapes, stone fruits such as peaches, nectarines, plums, apricots, prunes, persimmons, and cherries, subtropical crops such as mangoes, avocados, olives, citrus fruits, oranges, lemons, limes, tangerines, grapefruits, kiwis, tangelos, kumquats, calamansi, java pomelo, figs, dates, and nuts such as almonds, pecans, pistachios, walnuts, filberts, hazelnuts, and chestnuts. Arable crop plants include grain crops such as wheat, maize (corn), rice, barley, millet, oats, and rye; pulse crops such as lentils, beans, and peas; oilseed crops such as rapeseed, soybeans, sunflowers, and linseed; fiber crops such as cotton, jute, and flax; vegetable crops such as tomatoes, peppers, potatoes, pumpkins, garlic, onions, leeks, carrots, celery, sugar beets, beets, spinach, lettuce, clover, cabbage, Brussels sprouts, broccoli, cauliflower, turnips, and cucumbers; fruit crops such as melons, watermelons, strawberries, raspberries, blueberries, pineapples, and bananas; flower crops such as lilies, orchids, and tulips; and other crops such as peanuts, sugarcane, cocoa, coffee, and tea.

[0054] More preferred crops are crops planted in rows, such as sugarcane, corn (maize), soybeans, rapeseed, sunflower, cotton, potato, tomato, pepper, cucumber, onion, wheat, barley, rice, grapes, peanuts, bananas, cocoa, and coffee.

[0055] In one example, the width (distance) between two rows of planted crops is 10 cm to 2 m, and more preferably 50 cm to 1.5 m.

[0056] In one example, the height of the crop is 20 cm to 10 m, preferably 20 cm to 5 m, and more preferably 50 cm to 3.5 m.

[0057] According to one example, the at least one image includes a plurality of images, and the control processing unit is configured to control the camera to capture the plurality of images at positions corresponding to a plurality of different positions relative to the ground within the height of the crop but below the vertical height of the crop and / or between rows of the crop but below the vertical height of the crop.

[0058] According to one example, the control processing unit is configured to analyze at least one image to determine a weed map that classifies the presence of weeds on the ground based on the density and spatial distribution of the weeds.

[0059] In one example, at least one image acquired from a UAV above the crop height can also be used to determine the weed map.

[0060] In one example, the control processing unit is configured to determine a weed map by analyzing at least one image that classifies the presence of weeds on the ground based on the density and spatial distribution of the weeds and based on the type of weed.

[0061] For example, weed density is defined as the density of a plant per square meter. 2 Refers to the number of weed plants per square meter.

[0062] In one example, the spatial distribution of weeds refers to the density of weeds at a particular location on a field. For example, the spatial pattern of weeds can be scattered (single individual plants randomly distributed across the field) or clumped (weeds form patches across the field). In addition, the spatial distribution of weeds can be densely and uniformly dispersed, for example, between rows of multiple crops.

[0063] In one example, the size of the UAV is approximately 1 cm to 20 cm, preferably approximately 2 cm to 10 cm, and more preferably approximately 4 cm to 8 cm, and the weight of the UAV is less than 200 g, preferably less than 100 g, and more preferably less than 50 g.

[0064] According to one example, the UAV includes a propeller guard 40. The propeller guard is configured to at least partially surround a propeller of the UAV, and preferably is configured to completely surround the propeller.

[0065] In one example, the propeller guard may be formed from a strong, lightweight material and may be molded into a three-dimensional shape or assembled from thin strips of material into a three-dimensional structure.

[0066] In one example, the propellers of the UAV are counter-rotating propellers.

[0067] In one example, the camera is mounted on an extendable and retractable mount on the UAV, the mount configured to vary the distance between the camera and the UAV body.

[0068] In one example, the camera is mounted on an extendable and retractable telescoping mount on the UAV.

[0069] In one example, the UAV includes a distance sensor configured to measure the distance between the extendable and retractable mount member and the weed.

[0070] In one example, the control processing unit of the UAV is configured to use distance sensor information to avoid collisions of the UAV and its extendable and retractable mounts with weeds and / or crop plants.

[0071] In one example, the UAV preferably includes a distance sensor selected from the group consisting of a LiDAR sensor, a parallax laser ranging sensor, a stereo vision sensor, an IR reflectance sensor, a time-of-flight sensor, an ultrasonic sensor, and a radar sensor.

[0072] In one example, the UAV includes a LiDAR sensor as a distance sensor, which is used to guide the UAV within spatially confined areas, such as within the height of crop plants, and to assist in collision avoidance.

[0073] According to one example, the UAV includes at least one weed control unit 50. The at least one weed control unit is configured to be activated at a location determined by the control processing unit based on image analysis regarding the presence and location of at least one weed.

[0074] In one example, the at least one weed control unit includes at least one spray unit configured to spray a liquid.

[0075] In one example, the spraying unit is positioned (or may be positioned using an extendable mounting member) in front of the downdraft from the UAV's propellers to avoid interference.

[0076] In one example, the term "activation" in the context of a spray unit refers to the initiation of the spray process.

[0077] In one example, the spray unit includes at least one liquid atomizer, such as a hydraulic nozzle, and / or at least one atomizing disk, such as a spinning disk.

[0078] In one example, at least one weed control unit includes a liquid atomizer, a liquid tank, and at least one supply pipe. The liquid tank is configured to contain a chemical. The supply pipe is configured to transfer the liquid from the liquid tank to the liquid atomizer. The liquid atomizer is configured to spray the liquid.

[0079] In one example, the liquid capacity of the liquid tank is 0.1 ml to 800 ml, and preferably 0.5 ml to 100 ml.

[0080] In one example, the term "liquid(s)" refers to liquid(s) that include chemical-based and / or biological-based agricultural herbicide(s).

[0081] In one example, the term "liquid(s)" also includes one or more adjuvants to be applied by at least one spray unit to enhance retention and biological delivery of the chemical-based and / or biological-based agricultural herbicides to target weeds. The adjuvants can be combined with the chemical-based and / or biological-based agricultural herbicides prior to or within the spray unit.

[0082] In one example, the control processing unit is configured to control the UAV to fly at a flight height of less than 0.8 meters (more preferably less than 0.4 meters) above the ground when the at least one spraying unit is activated.

[0083] In one example, the control and processing unit is configured to activate at least one spray unit to dispense liquid either as a spray of fine droplets, as a single jet, as a single droplet, or as a combination thereof, depending on the preferred type of deposit.

[0084] In one example, the control processing unit is configured to adjust the amount of liquid to be sprayed onto the target by the at least one spray unit depending on the size of the target.

[0085] In one example, the control processing unit is configured to control and fly the UAV and to activate at least one spraying unit to achieve band spraying, patch spraying, and inter-target spraying, or combinations thereof, on agricultural land.

[0086] In one example, the control processing unit is configured to control and fly the UAV and to activate at least one spraying unit to spray liquid in a spray distribution pattern inside the crop to minimize the spread of weeds.

[0087] In one example, the at least one weed control unit is mounted on the UAV on at least one extendable and retractable mount member, the at least one mount member configured to vary the distance between the at least one weed control unit and the UAV body.

[0088] In one example, the at least one weed control unit is mounted on the UAV on at least one extendable and retractable telescoping mount member.

[0089] In one example, the UAV includes a distance sensor configured to measure the distance between the extendable and retractable mount member and the weeds. Useful distance sensors are described in detail herein above.

[0090] In one example, the control processing unit of the UAV is configured to use distance sensor information to prevent the UAV and its extendable and retractable mounts from colliding with the weed target.

[0091] According to one example, the UAV includes a position determining means 80 .

[0092] The position determining means 60 is configured to provide the control processing unit 20 with at least one position associated with the camera when the at least one image related to the ground was acquired.

[0093] The location can be a geographic location relative to an exact location on the ground, or it can be a location on the ground relative to one or more other locations on the ground, such as a field boundary or a base station location. In other words, an absolute geographic location can be used, or a location on the ground relative to a known location, although not necessarily known in absolute terms, can be used.

[0094] In one example, the location is an absolute geographic location.

[0095] In one example, the location is determined with reference to one or more known locations.

[0096] In other words, an image can be determined to be associated with a particular location on the ground without knowing its exact geographic location, but knowing the location where the image was captured relative to one or more known locations on the ground allows the location where the image was captured to be recorded. In other words, an absolute GPS-derived location where the UAV captured an image on the ground can be provided and / or the location where the image was captured relative to a known location, such as a field boundary or the location of a base station for the UAV, and again, the control processing unit can determine the exact location where the image was captured because the absolute location of the field boundary or charging station is known.

[0097] In one example, a GPS unit is used to determine and / or in determining a location, such as a camera position, when a particular image was acquired.

[0098] In one example, an inertial navigation unit is used alone or in combination with a GPS unit to determine a position, such as a camera position, when a particular image was acquired. Thus, for example, an inertial navigation unit including one or more laser gyroscopes can be calibrated or zeroed at a known location (e.g., a base station) and, when moving with at least one camera, can determine movement in x, y, and z coordinates away from that known location and, based thereon, determine the position of the at least one camera when the image was acquired.

[0099] According to one example, the UAV includes a transceiver (70). The control processing unit is configured to utilize the transceiver to transmit information regarding at least one image associated with the ground and / or information regarding image analysis of the at least one image associated with the ground, as well as at least one location associated with the camera when the at least one image associated with the ground was acquired. The control processing unit is configured to receive instructions for flying the at least one weed to the determined at least one location and to receive instructions for activating the at least one weed control unit 50 at the at least one location. The instructions are based on the at least one image associated with the ground and / or image analysis of the at least one image associated with the ground transmitted by the transceiver.

[0100] Thus, the UAV acquires images (optionally along with basic image analysis) that are transmitted to one or more processors located external to the UAV, which perform detailed image analysis. The external processor may receive images from multiple different UAVs (and locations) and can calculate multiple flight targets and optimal flight paths for identified flight targets while minimizing interference between adjacent UAVs. Flight path information is provided to individual UAVs that use this information to fly to identified targets and activate at least one weed control unit.

[0101] In one example, the UAV includes a multispectral light source that can aid in weed identification (interior to the vegetation, where less light is available) and navigation during flight, including at dawn, dusk, and night.

[0102] In one example, the UAV includes at least one additional sensor, such as a thermal camera, a sensor for measuring soil moisture, and a sensor for measuring soil temperature. Data from such sensors is useful in planning weed control operations because such factors affect weed and crop growth. Other sensors may be associated with the UAV, such as a direction sensor, an orientation sensor, an elevation sensor, a battery power sensor, and a position sensor relative to the base station, a sound sensor, etc.

[0103] FIG. 2 illustrates an example of a base station 100 for multiple UAVs 10. The base station includes a UAV parking unit 110 including at least one UAV 10, a control processing unit 120, and at least one transceiver 130. The at least one UAV includes a control processing unit 20, a camera 30, and a transceiver 70. The UAV parking unit is configured to transport multiple UAVs. The control processing unit 20 is configured to fly the UAV from the base station to a position within the height of the crops but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. The control processing unit 20 is configured to control the camera to capture at least one image related to the ground within the height of the crops but below the vertical height of the crops and / or between rows of the crops but below the vertical height of the crops. Control processing unit 20 is configured to transmit information relating to at least one image related to the ground to a base station using transceiver 70. Control processing unit 120 is configured to receive information relating to at least one image related to the ground from at least one UAV using transceiver 130. Control processing unit 120 is configured to determine the presence and location of at least one weed on the ground by analyzing the at least one image received from the at least one UAV.

[0104] In one example, the base station may be an unmanned ground vehicle (UGV), a transport vehicle, a UAV, a tractor, or the like.

[0105] In one example, the base station is a UAV delivery vehicle.

[0106] In one example, a UAV delivery vehicle typically has a size of 0.5 m to 2 m and a payload capacity of several kg, preferably greater than 2 kg, allowing it to carry, for example, 40 UAVs weighing 50 g each, or 80 UAVs weighing 25 g each. The delivery vehicle can carry a mixture of UAVs of different sizes for different purposes (e.g., UAVs of different sizes for agricultural land reconnaissance and specialized UAVs for weed control).

[0107] In one example, the UAV delivery vehicle includes at least one leg, preferably multiple legs. In one example, the (at least one) leg is extendable. Thus, the UAV delivery vehicle can land, for example, in the center or on the edge of a crop field without disturbing the growing crops.

[0108] In one example, image analysis is performed by the processing unit 120 of the base station 100. Since there are no size limitations as with UAVs flying inside the crop canopy, the base station can have extensive processing power and analyze images from multiple UAVs, in a manner similar to that described above.

[0109] In one example, the base station includes a service unit configured to recharge the power unit (battery), refill at least one liquid tank with liquid (if available), replace at least one liquid tank with an at least partially filled liquid tank, clean the spray unit, and / or provide other necessary services for the UAV.

[0110] According to one example, base station 100 includes a plurality of UAVs 10. The UAVs further include position determining means 60. The position determining means 60 is configured to provide control processing unit 20 with at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 20 is configured to transmit, using transceiver 90, information regarding the at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 120 is configured to receive, using transceiver 130, information regarding the at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 120 is configured to analyze at least one image received from at least one UAV to determine the presence of at least one weed on the ground at the at least one position.

[0111] According to one example, the base station 100 includes a plurality of UAVs 10. At least one UAV further includes at least one weed control unit 50. The control processing unit 120 is further configured to determine instructions for flying the at least one UAV to the at least one location where the presence of at least one weed has been determined, and further to determine instructions for activating the at least one weed control unit at the at least one location.

[0112] In one example, the control processing unit 120 is configured to analyze the history of previous liquid treatments in the field to select a liquid suitable for spray application with a different mechanism of action than the immediately preceding treatment, thereby minimizing the development of resistance, determine instructions for flying at least one UAV to the at least one location where the presence of at least one weed has been determined, and further determine instructions for activating at least one spraying unit at the at least one location.

[0113] In one example, the control processing unit 120 is configured to identify locations where the liquid is not working effectively, to select a liquid suitable for high-resistance spray application, to determine instructions for flying at least one UAV to the at least one location, and to determine instructions for activating at least one spray unit at the at least one location.

[0114] In one example, the control processing unit 120 is configured to identify UAVs that fail to spray liquid at the target locations, determine instructions for flying at least one other UAV to the at least one target location, and further determine instructions for activating at least one spraying unit at the at least one target location.

[0115] According to one example, base station 100 for multiple UAVs includes UAV parking unit 110 housing multiple UAVs. Control processing unit 120 is configured to determine a weed map by analyzing at least one image received from at least one UAV. The weed map classifies the presence of weeds on the ground based on the density and spatial distribution of the weeds (and preferably also based on the type of weed). Control processing unit 120 is further configured to determine a flight path map for the multiple UAVs as a function of the weed map and as a function of the weed classification based on the density and spatial distribution of the weeds on the ground. Control processing unit 120 is further configured to determine instructions for flying the multiple UAVs to the at least one location where the presence of at least one weed has been determined, and further to determine instructions for activating at least one weed control unit 50 at the at least one location based on the determined flight path map.

[0116] In one example, the flight path map includes information regarding activation and deactivation of at least one weed control unit of the UAV.

[0117] In one example, the flight path map and the activation / deactivation of at least one weed control unit along the flight path map of the UAV include information regarding swath weed control operation activity, patch weed control operation activity, and inter-target weed control operation activity, or combinations thereof.

[0118] According to one example, the base station 100 for multiple UAVs 10 includes UAVs, and the control processing unit 120 is configured to control the multiple UAVs to operate as alternating groups, i.e., when one group is in flight, another group is on the UAV parking unit.

[0119] In one example, a UAV on a parking unit is served by a service unit in a base station.

[0120] In one example, the parking unit is an open landing bay configured to allow multiple UAVs to take off and land.

[0121] In one example, the parking unit is configured to park multiple UAVs and to lock / unlock the multiple UAVs, where locking of the UAVs is required, for example, when the base station is moved.

[0122] In one example, a base station 100 for multiple UAVs includes at least one UAV with a multispectral camera configured to perform observation of the agricultural land prior to takeoff of the UAV(s) 10. The observation includes determining reconnaissance targets. The observation further includes calculating how to effectively position the UAV 10 to manage the agricultural land.

[0123] In one example, a base station 100 for multiple UAVs includes at least one recovery UAV configured to recover any UAV 10 that is unable to return to the base station under its own power. The recovery UAV may include a robotic arm configured to return any UAV 10 to the base station.

[0124] 3 illustrates a system 200 for agricultural weed management. The system 200 includes a base station 100 for multiple UAVs 10. The base station 100 includes a UAV parking unit 110 including at least one UAV, a control processing unit 120, and at least one transceiver 130. The at least one UAV 10 includes a control processing unit 20, a camera 30, and a transceiver 70. The UAV parking unit is configured to transport the multiple UAVs. The control processing unit 20 is configured to fly the UAV from the base station to a position within the height of the crops and below the vertical height of the crops, and / or to a position between rows of the crops and below the vertical height of the crops. Control processing unit 20 is configured to control the camera to capture at least one image of the ground within the height of the crop but below the vertical height of the crop and / or between rows of the crop but below the vertical height of the crop. Control processing unit 20 is configured to transmit information regarding the at least one image of the ground to a base station using transceiver 70. Control processing unit 120 is configured to receive information regarding the at least one image of the ground from at least one UAV using transceiver 130. Control processing unit 120 is configured to determine the presence and location of at least one weed on the ground by analyzing the at least one image received from the at least one UAV.

[0125] In one example, system 200 includes base station 100 for at least one UAV 10. The UAV further includes position determining means 60. Position determining means 60 is configured to provide control processing unit 20 with at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 20 is configured to transmit, using transceiver 70, information regarding the at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 120 is configured to receive, using transceiver 130, information regarding the at least one position associated with the camera when at least one image associated with the ground was acquired. Control processing unit 120 is configured to analyze at least one image received from the at least one UAV to determine the presence of at least one weed at the at least one position.

[0126] In one example, system 200 includes a base station 100 for a plurality of UAVs 10. At least one UAV further includes at least one weed control unit 50. Control processing unit 120 is further configured to determine instructions for flying the at least one UAV to the at least one location where the presence of at least one weed has been determined, and further to determine instructions for activating the at least one weed control unit at the at least one location.

[0127] 4 illustrates an example method for agricultural weed management 300. The method includes a flying step 310, also referred to as step a), in which the UAV is flown into a position within the height of the crop and below the vertical height of the crop, and / or between rows of multiple crops and below the vertical height of the multiple crops.

[0128] In an acquisition step 320, also referred to as step b), the camera acquires at least one image related to the ground within the height of the crop and below the vertical height of the crop and / or between rows of multiple crops and below the vertical height of the multiple crops.

[0129] In an analysis step 330, also referred to as step c), at least one image is analyzed to determine at least one weed and its location on the ground.

[0130] In one example, step b) comprises acquiring a plurality of images, wherein the control processing unit is configured to control the camera to acquire a plurality of images corresponding to a plurality of different positions relative to the ground within the height of the plants and below the vertical height of the plants and / or between rows of the plants and below the vertical height of the plants.

[0131] In one example, step c) comprises analyzing at least one image to determine a weed map that classifies the presence of weeds on the ground based on density and spatial distribution of the weeds.

[0132] In one example, in step a), the UAV is flown to a position within the height of the crop and below the vertical height of the crop, and / or between rows of multiple crops and below the vertical height of the multiple crops, wherein the size of the UAV is about 1 cm to 20 cm, preferably about 2 cm to 10 cm, more preferably about 4 cm to 8 cm, and the weight of the UAV is less than 200 g, preferably less than 100 g, more preferably less than 50 g.

[0133] In one example, step a) involves flying the UAV to a position within the height of the crop and below the vertical height of the crop, and / or between rows of multiple crops and below the vertical height of the multiple crops, wherein the UAV includes a propeller guard configured to at least partially surround a propeller of the UAV, preferably configured to completely surround the propeller.

[0134] In one example, activation step 340, also referred to as step d), activates at least one weed control unit of the UAV at a location determined by the control processing unit based on image analysis of at least one image related to the ground.

[0135] In one example, in step a), the UAV is flown to a position within the height of the crop and below the vertical height of the crop, and / or between rows of multiple crops and below the vertical height of the multiple crops, in which case the UAV includes a position determination means; The acquisition step 321, also referred to as step b1), involves acquiring at least one image related to the ground surface and providing to the control processing unit at least one position associated with the camera when the at least one image related to the ground surface was acquired.

[0136] In one example, in step a), flying the UAV to a position within the height of the crop and below the vertical height of the crop, and / or between rows of the plurality of crops and below the vertical height of the plurality of crops, wherein the UAV includes a transceiver and a position determining means; The acquiring step 322, also referred to as step b2), comprises acquiring at least one image related to the ground surface and providing to the control processing unit at least one position associated with the camera when the at least one image related to the ground surface was acquired; A transmitting step 323, also referred to as step b3), comprises transmitting information about the at least one image related to the ground surface and / or information about an image analysis of the at least one image related to the ground surface, as well as at least one position associated with the camera when the at least one image related to the ground surface was acquired, In a receiving step 324, also referred to as step b4), instructions for flying at least one weed to the determined at least one location and for activating at least one weed control unit at the at least one location are received, where the instructions are based on at least one image related to the ground transmitted by the transceiver and / or image analysis of the at least one image related to the ground.

[0137] FIG. 5 shows a schematic diagram of a detailed example of a base station and multiple UAVs on a field. In this illustration, the base station is a UAV carrier vehicle carrying multiple small UAVs. The carrier vehicle can fly the UAVs to the field, land them, park them, or hover them. After arriving at the field, the UAV carrier vehicle can launch the small UAVs, which can fly across the field and detect weeds on the ground using an optimal flight pattern. The UAVs can locally spray herbicides as needed. The small UAVs fly both above and within the crops, or between rows of crops such as vineyards, identifying weeds on the ground and controlling them by locally spraying herbicides. On the right side of FIG. 5, a UAV (equipped with a camera and a propeller protected by a propeller guard) captures images of weeds on the ground. Two small UAVs in the center of the figure (each equipped with a spraying unit and having a propeller protected by a propeller guard) spray herbicides on weeds on the ground by spot spraying. The control processing unit can determine the appropriate herbicide to be sprayed based on information received from the camera and related sensors (e.g., depending on the type of weed detected). A particular advantage of the present invention is that a group of UAVs can very quickly perform reconnaissance and autonomously spray herbicides on multiple areas of the farmland, thereby enabling the entire farmland to be treated in a relatively short time. The operation speed depends on the number of UAVs: a group of 20 UAVs can simultaneously treat 20 different areas within the target farmland, while a group of 40 UAVs can treat 40 different areas, making the operation speed twice as fast. A further advantage of the present invention is that the field can be treated quickly by applying the herbicide only to the target areas, thus minimizing application to non-target areas.This is a highly effective method for targeting herbicides, significantly reducing the amount of herbicide applied to the field without reducing efficacy. This has environmental benefits and reduces residues in the crop. This advantage also reduces the amount of herbicide that needs to be delivered by the application device, from several liters per hectare to approximately 100 ml / hectare to 10 ml / hectare, an amount that can be easily delivered by a fleet of UAVs. Furthermore, because weed problems occur continuously throughout the growing season, an advantage of the present invention is that the field can be quickly revisited for repeated treatments, for example, every 5 to 10 days, ensuring that target weeds are treated at their optimal growth stage (e.g., 2-leaf to 5-leaf stage). Additionally, early-stage weeds are easier to control than later-stage weeds, reducing the amount of active ingredient required. Additionally, successful biological control of weeds often requires careful application windows, and field application with tractor-mounted boom sprayers is costly, including driver time, consumes a lot of energy and generates correspondingly high CO2 emissions, and can cause compaction damage to the soil, especially if treatments must be repeated when new weeds emerge.

[0138] 6a-6c show schematic examples (plan view) of multiple UAVs spraying herbicides on a crop field based on different weed maps. In the example of FIG. 6a, the density and spatial distribution of identified weeds on the field dictate a patch-like spraying pattern. In this example, a UAV flies to identified weed patches within a row of crops (shown as black dots) and sprays herbicides only on the weeds within the weed patch. In the next row, a second UAV flies to identified weed patches within this row and sprays herbicides as described above. In the third row, a third UAV operates similarly. In the example of FIG. 6b, the number of weeds is high, favoring a strip-like spraying strategy. In this case, for each row, the UAV activates a weed control unit to spray herbicides along the entire strip (row). In the example of Figure 6c), because the number of weeds is small, a target-to-target spraying strategy is advantageous, in which the UAV flies within the row and sprays the herbicide individually on the identified weeds within the row. These three strategies can also be combined depending on the weed growth conditions on the farm. Additionally, the UAV does not necessarily need to fly within the rows of multiple crops and spray the herbicide within the rows. Other flight paths and herbicide spraying paths are also possible, such as paths between different rows.

[0139] In another exemplary embodiment, a computer program or a computer program product is provided, characterized in that it is configured to perform, on a suitable system, each of the method steps of the method according to one of the previous embodiments.

[0140] Thus, the computer program product may be stored on a computing unit, which may also be part of an embodiment. This computing unit may be configured to perform or to guide the performance of the steps of the above-described method. Furthermore, it may be configured to operate the components of the above-described UAV, base station, and / or system. The computing unit may be configured to operate automatically and / or to execute user instructions. The computer program may be loaded into a working memory of a data processor. Thus, the data processor may be provided to perform the method according to any of the previous embodiments.

[0141] This exemplary embodiment of the present invention covers both computer programs that originally use the present invention and computer programs that have been updated to convert existing programs into programs that use the present invention. Furthermore, the computer program product may be capable of providing all the steps necessary to complete the procedures of the exemplary embodiments of the methods described above.

[0142] According to a further exemplary embodiment of the present invention, a computer readable medium such as a CD-ROM, a USB stick or the like is presented, which has stored therein a computer program product that is / can be a computer program product as described in the previous paragraph. The computer program can be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid-state medium, provided together with or as part of other hardware, but also in other forms, such as via the Internet or via other wired or wireless telecommunication systems.

[0143] However, the computer program may also be presented over a network, such as the World Wide Web, and can be downloaded from such a network into the working memory of a data processor.

[0144] According to a further exemplary embodiment of the present invention, a medium for making a computer program product downloadable is provided, the computer program product being configured to perform a method according to one of the previously described embodiments of the present invention.

[0145] It should be noted that embodiments of the present invention are described with reference to different subject matters. In particular, some embodiments are described with reference to method-type claims, while other embodiments are described with reference to UAV, base station, and / or system-type claims. However, a person skilled in the art will infer from the above and following descriptions that, unless otherwise specified, any combination of features belonging to one type of subject matter, as well as any combination of features related to different subject matters, is considered to be disclosed by the present application.

[0146] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the dependent claims.

[0147] In the claims, the word "comprising" does not exclude other elements or other steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be interpreted as limiting their scope.

Claims

1. An unmanned aerial vehicle (hereinafter "UAV") (10) for agricultural weed management, comprising: a control processing unit (20), a camera (30), the control processing unit is configured to control the UAV to fly to a position within a crop height and below a vertical height of the crop, and / or to a position between rows of a plurality of crops and below a vertical height of the plurality of crops; the control processing unit is configured to control the camera to capture at least one image related to ground at a position within the height of the plant and below the vertical height of the plant, and / or at a position between the rows of the plurality of plant species and below the vertical height of the plurality of plant species; the control processing unit is configured to determine the presence and location of at least one weed on the ground by analyzing the at least one image; The control processing unit of the UAV (10) is configured to analyze the at least one image to determine a weed map that classifies the presence of weeds on the ground based on the density and spatial distribution of the weeds.

2. 2. The UAV of claim 1, wherein the at least one image includes a plurality of images, and the control processing unit is configured to control the camera to acquire a plurality of images at positions corresponding to a plurality of different positions relative to the ground within the height of the plant but below the vertical height of the plant and / or between the rows of the plant but below the vertical height of the plant.

3. A UAV as described in claim 1 or 2, wherein the size of the UAV is approximately 1 cm to 20 cm, preferably approximately 2 cm to 10 cm, and more preferably approximately 4 cm to 8 cm, and the weight of the UAV is less than 200 g, preferably less than 100 g, and more preferably less than 50 g.

4. The UAV is - includes a propeller guard (40), 4. A UAV as described in any one of claims 1 to 3, wherein the propeller guard is configured to at least partially surround a propeller of the UAV, preferably configured to completely surround the propeller.

5. The UAV is - comprising at least one weed control unit (50), 5. A UAV as described in any one of claims 1 to 4, wherein the at least one weed control unit is configured to be activated at a location determined by the control processing unit based on the image analysis regarding the presence and location of the at least one weed.

6. The UAV includes a position determining means (60); 6. The UAV of claim 1, wherein the position determining means (60) is configured to provide the control processing unit (20) with at least one position associated with the camera when the at least one image relative to the ground was acquired.

7. the UAV includes a transceiver (70); the control processing unit is configured to transmit, by utilizing the transceiver, information regarding the at least one image related to a ground surface and / or information regarding image analysis of the at least one image related to a ground surface, and at least one position associated with the camera when the at least one image related to a ground surface was acquired; the control processing unit is configured to receive instructions for flying at least one weed to the determined at least one location and to receive instructions for activating the at least one weed control unit (50) at the at least one location; The UAV of claim 5, wherein the instructions are based on the at least one image related to the ground transmitted by the transceiver and / or the image analysis of the at least one image related to the ground.

8. A base station (100) for a plurality of UAVs, comprising: The base station a UAV parking unit (110) containing at least one UAV; a control processing unit (120), at least one transceiver (130), The at least one UAV a control processing unit (20), a camera (30), a transceiver (70), the UAV parking unit is configured to transport a plurality of UAVs; the control processing unit (20) is configured to control the UAV to fly from the base station to a position within a crop height and below the vertical height of the crops, and / or to a position between rows of a plurality of crops and below the vertical height of the plurality of crops; the control processing unit (20) is configured to control the camera to acquire at least one image related to the ground at a position within the height of the plant and below the vertical height of the plant, and / or at a position between the rows of the plurality of plant species and below the vertical height of the plurality of plant species; the control processing unit (20) is configured to transmit information about the at least one image related to the ground to the base station by utilizing the transceiver (70); the control processing unit (120) is configured to receive, by utilizing the transceiver (130), information relating to the at least one image associated with the ground from the at least one UAV; The control processing unit (120) is configured to determine the presence and location of at least one weed on the ground by analyzing the at least one image received from the at least one UAV, and to determine a weed map that classifies the presence of the weeds on the ground based on the density and spatial distribution of the weeds.

9. The at least one UAV further includes a position determining means (60); the position determining means (60) is configured to provide the control processing unit (20) with at least one position associated with the camera when the at least one image relative to the ground was acquired; the control processing unit (20) is configured to transmit, by utilizing the transceiver (70), information relating to the at least one position associated with the camera when the at least one image related to the ground was acquired; the control processing unit (120) is configured to receive, by utilizing the transceiver (130), the information regarding the at least one position associated with the camera when the at least one image related to the ground was acquired; 9. The base station for a plurality of UAVs of claim 8, wherein the control processing unit (120) is configured to determine the presence of at least one weed on the ground at the at least one location by analyzing the at least one image received from the at least one UAV.

10. the at least one UAV further comprises at least one weed control unit (50); 10. A base station for a plurality of UAVs as described in claim 8 or 9, wherein the control processing unit (120) is further configured to determine instructions for flying the at least one UAV to the at least one location where the presence of the at least one weed has been determined, and further to determine instructions for activating the at least one weed control unit (50) at the at least one location.

11. a UAV parking unit (110) having a plurality of UAVs; the control processing unit (120) is configured to determine a weed map classifying the presence of weeds on the ground based on weed density and spatial distribution (and preferably also based on weed type) by analyzing the at least one image received from the at least one UAV; the control processing unit (120) is further configured to determine a flight path map for the plurality of UAVs as a function of the weed map and as a function of the classification of weeds based on density and spatial distribution of weeds on the ground; 11. A base station for a plurality of UAVs as described in claim 10, wherein the control processing unit (120) is further configured to determine instructions for flying the plurality of UAVs to the at least one location where the presence of the at least one weed has been determined, and further to determine instructions for activating the at least one weed control unit (50) at the at least one location based on the determined flight path map.

12. 12. A base station for a plurality of UAVs as described in any one of claims 8 to 11, wherein the control processing unit (120) is configured to control the plurality of UAVs to operate in alternating groups, i.e., when one group is in flight, another group is on the UAV parking unit.

13. A method (300) for agricultural weed management, comprising: a) flying a UAV to a position within the height of a crop and below the vertical height of the crop, and / or between rows of a plurality of crops and below the vertical height of the plurality of crops; b) an acquisition step (320) of at least one image related to the ground by a camera at a position within the height of the plant and below the vertical height of the plant and / or at a position between the rows of the plant and below the vertical height of the plant; c) an analysis step (330) of determining at least one weed and its location on the ground by analyzing the at least one image, and determining a weed map that classifies the presence of the weeds on the ground based on the density and spatial distribution of the weeds.

14. 8. A computer program product for controlling a UAV according to any one of claims 1 to 7, comprising: A computer program product configured to perform the method of claim 13 when executed by a processor.

Images