Applicator, system and method for applying electrical current into a plant, and agricultural pivot using a tethered drone

The tethered drone applicator system addresses processing and transmission limitations in electrical weeding by using computer vision for real-time weed detection and treatment, enhancing precision and sustainability in agricultural applications.

WO2026232205A1PCT designated stage Publication Date: 2026-11-12ZASSO GRP AG +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/BR2025/050164
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-05
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing electrical weeding systems face challenges such as high processing requirements, signal transmission failures, and high costs due to the need for powerful processors and complex image processing, limiting their efficiency and environmental impact.

Method used

A tethered drone applicator system with an image sensor, electrode, and vertical mover for precise application of electrical current to undesired plants, integrated with agricultural pivots, utilizing computer vision for real-time detection and treatment.

Benefits of technology

Enables precise and efficient weed control with reduced environmental impact by minimizing processor requirements and signal transmission challenges, while integrating with existing agricultural infrastructure for large-scale operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BR2025050164_12112026_PF_FP_ABST
    Figure BR2025050164_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to an electrical weeding device designed to enhance precision agriculture by integrating advanced electrical weeding technology with drone mobility and commonly used carriers such as pivots or tractors. The device comprises at least one electrical converter to provide the necessary current parameters for effective weeding, at least one sensor configured to identify and locate target plants, and at least one drone carrying at least one electrode for applying electrical current to the undesired plants. This electrode is connected to a drone that is tethered to a carrier, enabling the drone to move the electrode towards the identified weeds upon detection. The device may feature a docking station mounted on the carrier, which can be an irrigation pivot, a tractor, a planter, or a sprayer. One electrode may be positioned at the tip of an insulated structure connected to the drone, while the other electrode is grounded elsewhere, ensuring efficient current flow for weeding. The operation of the device may be structured into three layers: Physical Layer: Comprising components such as the drone, vision sensor, actuator, carrier, electrode, and tether. Logic Layer: Including elements such as a processing unit for sensor data, flight control, computation of drone actions, a database for gathering information, and georeferencing of targeted plant locations. Application Layer: Encompassing processes for flight planning, mission control, operation of electrodes or actuators, and flight data logging. This layered architecture ensures the device's effective functionality, maintenance, and scalability, providing an innovative solution for targeted weed management in agricultural settings.
Need to check novelty before this filing date? Find Prior Art

Description

APPLICATOR, SYSTEM AND METHOD FOR APPLYING ELECTRICAL CURRENT INTO A PLANT, AND AGRICULTURAL PIVOT USING A TETHERED DRONE FIELD OF THE DISCLOSURE

[0001] The present invention relates to an electrical current applicator for the application of electricity into plants comprising an image sensor and a mover for moving an electrode upon detection of undesired plants in crops such as weeds. More specifically, the invention refers to an electrical weeding tethered drone home applicator device that can be used in structures of crop fields such as pivot irrigation systems.DESCRIPTION OF THE RELATED ART

[0002] With the advancement of technologies in agriculture, electrical weeding devices have emerged as an efficient solution for eliminating weeds. These devices use a combination of sensors, image processing systems, and methods to apply electrical energy to treat unwanted plants with precision, minimizing the use of chemical herbicides and contributing to more sustainable agricultural practices.

[0003] Document US10548306B2 discloses a device for damaging uncultivated plants, comprising a tool which is designed to damage the uncultivated plants and a classification unit representing positional data of the uncultivated plants and / or designed to recognize uncultivated plants. The classification unit has the positional data of the weed available, especially in the form of a map, and / or if it detects the weed with the aid of a camera unit and / or a laser unit. This measure makes it possible to further increase the precision of the weed-control process.

[0004] Document W02017002093A1 refers to a method of using a robot for an automatic treatment of a weed, by capturing images of weeds with a camera on said robot; determining with said images a distance between the weed and a second vegetable organism; and selecting a treatment tool of saidrobot for a treatment of the weed based on the determined distance. Document W02017002093A1 describes a computer vision system as part of the processing system that processes the successive images captured with the camera and determine the distance between the vegetable organisms. The computer vision system can identify the species of the vegetable organism on the surface and map their position in the field, permitting an accurate treatment of the vegetable organism.

[0005] Document W02021069980A1 relates to a robotic weeding equipment for an agriculture robotic weeding system, comprising a plant squeezer for squeezing cultivated plants and an image recognition device for identification of weeds; and a weed treatment processing device. Thus, W02021069980A1 discloses an image recognition device that is capable of identifying weeds which may be located near the base of each rows of cultivated plants since their leaves are moved away from the field of vision of the image recognition device by respective longitudinal sides of the plant squeezer and weed treatment may be selected to eradicate the weeds.

[0006] Document US20170251589A1 discloses a farming system including a field engagement unit having a support assembly, most specifically, it discloses an agricultural pivot comprising several attachments for different purposes. The support assembly includes one or more work tool rail assemblies. The farming system additionally includes a local controller. The local controller includes one or more processors configured to execute a set of program instructions stored in memory. In addition, US20170251589A1 describes the use of vision systems to capture image of crops comprising crop features. In addition, the one or one crop features captured may be used for identifying the weed and proactively guiding a mechanical weeding and / or herbicide-spraying attachment. According to US20170251589A1, some of the attachments may be configured to mechanically weed around specific plants and / or regions of the field based on processed imaging recommendations.

[0007] Document AU2017366640 Al relates to a device and a method for introducing a high voltage into a substrate comprising biological material, in which an applicator module having two or more applicators with simultaneously differing polarities is used through which the metered electrical high voltage flows so as to alter the substrate. Different embodiments allow the high voltage to be controlled and introduced in a targeted manner into the substrate.

[0008] Document WO2015119523A1 discloses an apparatus for destroying weeds, which consists of a switch transformer, cascade assembly for the voltage multiplication, and high voltage transformer and a feedback assembly, assembled in a whole so that input connector is attached to switch, output connector of which is attached to one input connector of the transformer's primary, whose other primary end is connected to input connector, while the output connectors of transformer's secondary are connected to input connectors of cascade assembly whose output is connected to one input connector of the high- voltage transformer's primary, whilst the cascade assembly is through feedback assembly connected to the second input connector of the high- voltage transformer's primary, whereas the connectors of the high- voltage transformer's secondary are connected to the electrodes which produce a strong electromagnetic field.

[0009] With the increasing demand for precision agriculture, vision systems have become a critical component in electrical weeding devices. These systems leverage advanced image capture and processing technologies to identify, classify, and target undesired plants, enabling precise and efficient weed control. By combining sensors, computational algorithms, and real-time data processing, vision systems enhance the accuracy of weeding operations while minimizing the environmental impact. Despite their potential, existing systems face limitations such as high processing requirements, signal transmission challenges, and cost barriers, highlighting the need for continuedinnovation in this field.[OO1O] Document CN103348964 describes a smart weeding device and method of using it, where the system describes a weeding system that captures images of the land and performs image definition processing for plant definition through a cloud and uses radio frequency for data transmission. A limitation of this type of device is the possibility of failure in the transmission of signals between the equipment of the system. In addition, the image processing of this invention performs convolution directly after RGB image capture, requiring high processor power and more processing time.

[0011] Document CN 105469020 describes a system equipped with weed sensing and a spraying device. The processing of the image obtained by the system is carried out through the Super-Green Model that uses a highlighting equation that establishes weight two for the green band of the pixel subtracting the value of the two other remaining bands of weight one, in other words the pixel in grayscale is given by 2G-R-B. The gray image is passed through the binarization filter through the threshold established by the OTSU Method of setting the exclusion threshold. This invention has the same limitation as the previous one, the need for high processing capacity and does not perform convolution in the processing of its image, obtaining different results.

[0012] Document 2018US168141 describes a stand-alone equipment with an imaging system that determines which plant should be eliminated. The camera present in the device identifies the types of plants identified and whether it should be eliminated or not. This equipment presents the limitation of the requirement of a very high processing and storage capacity on the part of the system, making the equipment expensive.

[0013] Document CN113498639 describes a weeding equipment with image processing. The imaging system uses a camera to define where and how the intentionally planted commercial plants are with a camera located between the rows that defines the plants that are in that location and the number of them.This equipment, despite presenting a sensor for plant identification, its purpose and processing method are for identifying the type of plants and their counting. The equipment requires a high processing capacity, making its cost very high.

[0014] Document PCT / BR2024 / 050019, which directly precedes the present invention, describes a sensing system and method for the identification of invasive plants that effectively solves the limitations of the state of the art mentioned above. The sensing system for the identification of invasive plants that comprises a CCD camera "coupled charges" RGB "Red / Green / Blue Color Space" for image acquisition, connected to a microcontroller, via DCMI interface "Digital Camera Memory Interface", for embedded processing of image data and detection of areas that correspond to green color in the HSV color model "Hue / Saturation / Value Color Space". The device performs image acquisition and processing between 10 and 15 FPS "Frames per Second", when a green color area is detected, the device sends a binary signal via RS485 interface to the machine in which it is attached performs the programmed action for the determined time.

[0015] Document PCT / BR2025 / 050051 the novelty lies in the integration of a sensing and image processing system that not only identifies invasive plants, but also effectively differentiates between the plants grown in the planting rows and the weeds situated between these rows. This is achieved through the recognition of the green color in the captured images and the accurate identification of the planting rows.

[0016] Therefore, the advancements in electrical weeding devices highlight the increasing integration of technologies such as image processing, sensing systems, and innovative methods of applying electrical energy. These innovations address challenges in precision agriculture and sustainability, paving the way for more efficient and environmentally friendly management solutions.SUMMARY OF THE INVENTION

[0017] The present invention provides an applicator for applying electrical current into a plant, the applicator being connected to an electrical power source, the applicator comprising: an image sensor configured to capture image data of an area to be treated and connected to a processor configured to perform the detection method for detecting an undesired plant; an electrode for applying electrical current by contacting the undesired plant; and a vertical mover configured to vertically move the electrode; wherein, when an undesired plant is detected, the horizontal mover moves the applicator horizontally above the undesired plant and / or the vertical mover moves the electrode toward the undesired plant.

[0018] The present invention also refers to a system for applying electrical current into a plant comprising: at least two support structures positioned on separate locations of an area to be treated; a horizontal mover positioned on the supporting structures; an applicator movably placed on the horizontal mover and being configured to travel over the area to be treated, the applicator comprising: an image sensor configured to capture image data of the area to be treated and connected to a processor configured to perform a detection method for detecting an undesired plant; an electrode for applying electrical current by contacting the undesired plant; and a vertical mover configured to vertically move the electrode; wherein, when an undesired plant is detected, the horizontal mover moves the applicator horizontally above the undesired plant and / or the vertical mover moves the electrode toward the undesired plant.

[0019] Moreover, the present invention refers to a system for applying electrical current into a plant, designed to be integrated with an agricultural pivot or similar carrier structure.

[0020] The present invention also refers to a method for applying electrical current into a plant with a system, comprising the steps of: moving a applicator with a horizontal mover across an area to be treated; obtaining imagedata from the area to be treated with an image sensor; and upon detecting an undesired plant with the image sensor: moving the applicator over the undesired plant; moving an electrode toward the undesired plant with a vertical mover; and applying electrical current into the undesired plant.BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The invention is explained in greater detail below on the basis of figures. Shown therein are:

[0022] Figure 1 shows the layers of functioning of the invention.

[0023] Figure 2 shows an embodiment comprising a drone tethered to a pivot with an electrode approaching the target plant and a housing for the drone.

[0024] Figure 3 shows an embodiment comprising a drone tethered to an agricultural equipment that moves through the field.DETAILED DESCRIPTION OF THE INVENTION

[0025] The following description is based on preferred embodiments of the invention applied to an applicator for applying electrical current into a plant. However, the present invention is not limited to a particular embodiment, as it will be clear for a skilled person.

[0026] Figure 1 illustrates the layered structure of the invention, comprising the physical layer, logic layer, and application layer. The physical layer includes essential hardware components such as the drone, equipped with a vision sensor and an electrode, connected via a tether to the ground power system. The cable has an operational range of approximately 12 meters and is designed to withstand mechanical stress in the agricultural environment. These components perform field operations such as weed identification and elimination. The logic layer consists of processing units such as the control module and a georeferenced database dynamically built according to the prototype’s actuation. This map is updated at runtime to prevent applications in already treated and is only accessible by the onboard computer during operation, without remote access, ensuring real-time data synchronization anddecision-making. The application layer, includes high-level functionalities like mission planning, operation control, and flight data logging, enabling user intervention and supervision. This mission planning is automatically defined based on the output of the vision system, which segments green areas and detects crop rows to construct the actuation map. Parameter adjustments are only available during the configuration stage, with no real-time updates. This modular structure ensures operational efficiency and scalability in agricultural applications.

[0027] Figure 2 illustrates an embodiment of the system integrated with a pivot irrigation system. In this setup, the tethered drone operates alongside the, which serves as the support structure for the system. The electrical weeding system has a modular design and requires specific fixation points on the pivot structure, along with a dedicated connection point for energy supply. The drone is equipped with a vision sensor for weed detection and an electrode for applying electrical current to the target plant. The tether connects the drone to the power supply and control system housed in the pivot. The drone operates independently with autonomous control, while the position of the pivot irrigation system is monitored using GPS / GNSS to adjust the mission plan as needed. This configuration highlights the system’s ability to integrate with existing agricultural infrastructure, allowing precise weed management in large-scale fields while maintaining compatibility with irrigation operations.

[0028] Figure 3 illustrates an alternative embodiment where the tethered drone is integrated with an agricultural planter attached to a tractor. The drone, equipped with a vision sensor, operates in synchronization with the planting process, targeting and treating undesired plants in real-time. Communication between the drone and the docking base can be carried out wirelessly via radio or serially via the tether cable. However, the high speed of the planter can impact the drone’s actuation, reducing the number of points the system can treat. The tether connects the drone to the tractor’s power and control systems.This embodiment demonstrates the system's versatility by enabling simultaneous weeding and planting operations, optimizing efficiency and productivity in precision agriculture.

[0029] Generally, the present invention relates to an applicator for applying electrical current into a plant, the applicator being connected to an electrical power source, the applicator comprising: an image sensor configured to capture image data of the area to be treated and connected to a processor configured to perform a detection method for detecting an undesired plant; an electrode for applying electrical current by contacting the undesired plant; and a tethered drone configured to move the electrode; wherein, when an undesired plant is detected, the drone moves the electrode toward the undesired plant.

[0030] The present invention provides an applicator for applying electrical current to a plant, where the applicator is connected to an electrical power source. The applicator comprises:

[0031] An image sensor configured to capture image data of an area to be treated and connected to a processor configured to perform a detection method for identifying an undesired plant.

[0032] An electrode for applying electrical current by contacting the undesired plant. A drone carrying the electrode is configured to move the electrode to the targeted position.

[0033] In this system, a tethered drone positions the applicator over the undesired plant, eliminating the need for horizontal or vertical movers. The drone ensures precise placement of the electrode.

[0034] An image sensor configured to capture image data of the area to be treated and connected to a processor configured to perform a detection method for identifying an undesired plant. An electrode for applying electrical current by contacting the undesired plant.

[0035] When an undesired plant is detected, the drone positions the applicator directly above the plant, allowing the electrode to make contact andapply electrical current.

[0036] Moreover, the present invention refers to an agricultural pivot comprising a system for moving the tether and the drone through the target area. The present invention also refers to a method for applying electrical current to a plant with a system, comprising the steps of:

[0037] Using a tethered drone to move an applicator across an area to be treated.

[0038] Obtaining image data from the area to be treated with an image sensor.

[0039] Upon detecting an undesired plant with the image sensor: Positioning the applicator over the undesired plant using the drone. Moving an electrode toward the undesired plant through the drone, or a vertical mover.

[0040] Applying electrical current to the undesired plant.

[0041] In this configuration, the high-voltage power module is external and not carried by the drone. Instead, it is integrated into the support structures or the ground system, providing a safer and more efficient power delivery method for the applicator system. The tethered drone is responsible for positioning the active electrode, eliminating the need for separate horizontal or vertical movers, which become optional.

[0042] Tethered Drones are aerial mobile robots that operate with constant connectivity via a cable attached to a power source. This power supply system allows the robots to achieve greater autonomy by eliminating the need for frequent battery recharges, enabling long-duration missions (Gu et al., 2016). The most common tethered systems include three main components: the ground power module, the aerial power module, and the retraction system.

[0043] The system’s power is sourced from the 380V supply that powers the irrigation pivot itself. To handle potential power failures, an Uninterruptible Power Supply (UPS) is implemented as a backup system. The aerial power module, located on the drone, converts this energy to safe andstable levels using voltage converters (AC-DC or DC-DC) (Saravanakumar et al., 2023). Precise voltage regulation ensures protection of sensitive electronic components during operation.

[0044] The cables used in such systems must be selected based on their mechanical, electrical, and data transfer characteristics. They should support the required current without significant voltage drops, be lightweight, and flexible to avoid restricting the drone's movement (Fauzi & Rahim, 2022).

[0045] The use of computer vision in agriculture is becoming increasingly important for mapping planting regions, enabling better production management (Osco, Arruda, Gongalves, & Dias, 2021). Computer vision works in conjunction with an automatic remote sensing system, capturing area data that is then quickly analyzed by a computer. The integration of these systems allows for various applications, such as monitoring and detecting crop lines (Soares, Abdala, & Escarpinati, 2018).

[0046] Currently, computer vision systems are also being employed in the automation of mobile robots to perform tasks autonomously, enhancing efficiency and speeding up routine tasks related to precision agriculture. The use of unmanned aerial vehicles for image collection is presented by Sapkota et al. (2023) to reduce the use of pesticides for weed control through aerial mapping of the region, achieving up to a 26.2% reduction in herbicide use in crops. A more direct approach involving robots in agriculture is provided by Sethia et al. (2021), which uses a vision and image capture system on a terrestrial robot to identify and remove weeds near the main crop, achieving 99.5% detection accuracy with the help of neural networks.

[0047] Green area identification is generally used to segment planting images, aiming to separate plant regions from non-plant regions. Based on a captured image, non-crop line areas are identified and evaluated for green objects using methods such as green-on-brown. This segmentation is done through algorithms that aim to identify relevant features for weed detection inimages (Moldvai, Mesterhazi, Teschner, & Nyeki, 2024).

[0048] The green region detection proposed by Pulido-Rojas, Molina-Villa, & Solaque (2016) uses a dark chamber and lamps to achieve more uniform lighting, reducing environmental influence on segmented colors and achieving greater uniformity in the captured image's color for filtering. Moldvai et al. (2024) compare weed detection results using neural networks and computer vision, concluding that computer vision shows promising results in applications with fewer than a hundred images per category.

[0049] Crop Line Identification is an essential activity for monitoring and automating mobile robots intended to perform tasks in precision agriculture plantations. Automatic crop line detection emerges as an alternative to these problems since data collection and processing occur automatically. The images obtained through sensing are processed and used to feed a computer vision algorithm, which will then identify the existing crop lines. There are several approaches to this problem, with the Hough transform being one of the most commonly used to detect known shapes (Chen, Ma, Wang, & Li, 2021). An alternative for crop line detection is presented by Soares, Abdala, & Escarpinati (2018), applying the Hough transform to segmented overlapping images to avoid discontinuities between different captures at medium altitude, achieving more than 80% detection accuracy. To improve image segmentation by the Hough transform, Silva (2020) proposes using the Radon Transform to enhance the results obtained by the Hough transform, achieving better detection in images where the lines obtained by the Hough transform are discontinuous.

[0050] The architecture is delineated into three distinct layers: the application layer, the logic layer, and the physical layer. Each of these layers plays a crucial and interdependent role in ensuring the effective operation of the integrated system, which is intended for use in agricultural environments for tasks such as weed control.

[0051] The physical layer comprises the essential hardwarecomponents of the system, including the docking base, the tethered drone equipped with vision sensors (such as RGB / RGB-D cameras and LiDAR) for real-time plant detection, and an electrode assembly featuring a vertical actuator for precise positioning and contact with undesired plants. The drone additionally incorporates an onboard processing unit for immediate image analysis. The docking base, mounted on agricultural equipment like pivots or tractors, provides continuous power through a tether cable (380V AC with UPS backup) and enables data transmission between the drone and control systems. Complementary components include environmental sensors (ultrasonic and IMU) for navigation and obstacle avoidance, along with a grounding electrode to complete the electrical circuit. Together, these elements form the operational foundation for autonomous weed identification and targeted electrical treatment while maintaining compatibility with existing agricultural infrastructure.

[0052] The logic layer encompasses the system's processing and software components, including the flight controller, the onboard computer, and the database. This layer is fundamental for processing the collected data and controlling the drone. Computer vision algorithms, along with the flight controller firmware, ensure that the drone operates synchronously and efficiently. The system identifies unwanted plants by segmenting green regions in the captured images. Crop lines are detected over the segmentation, and green regions located between these lines are classified as invasive plants.

[0053] The identified green regions are initially referenced in image coordinates and are then transformed into the drone's movement space using a transformation frame (TF), aligning them with the local positioning reference system. The drone utilizes navigation sensors such as an IMU, barometer, accelerometer, gyroscope, and GPS for autonomous flight control. The PX4 flight controller receives the transformed coordinates of the green regions as a reference and directs the drone to the corresponding location, where theelectrical weeding treatment is applied.

[0054] Additionally, perception sensors including RGB and RGB-D cameras, LiDAR, and ultrasonic sensors contribute to precise localization and obstacle avoidance, ensuring accurate and efficient operation. The integration between the flight controller and the onboard computer ensures that all drone operations are synchronized safely.

[0055] The application layer includes the highest-level system elements, such as mission planning, control, operation, and flight records. Mission planning involves defining the objectives and trajectories the drone must follow, while control and operation ensure that the drone satisfactorily completes these missions. Additionally, flight records are maintained for later analysis, allowing system performance evaluation and identifying possible improvements. This layer ensures the possibility of user intervention, allowing the irrigation pivot operator to start and stop the prototype's operation at any time.

[0056] This layered structure provides a clear and functional organization of the different components and their respective responsibilities. The separation into layers facilitates the system's maintenance, updates, and scalability, ensuring it can evolve according to needs and technological innovations in the field of automation and robotics.

Claims

CLAIMS1. Electrical weeding device, characterized by the fact that it comprises:at least one electrical converter configured to adjust the current parameters to perform electrical weeding;at least one sensor configured to identify and locate a target plant; at least one electrode connected to the electrical converter, configured to apply electrical current by contacting the undesired plant; anda drone tethered to a carrier, configured to move the electrode toward the target plant upon detection.

2. Electrical weeding device according to claim 1, characterized by the fact that it comprises a docking station positioned in the carrier.

3. Electrical weeding device according to claim 1 or 2, characterized by the fact that the carrier being selected from an irrigation pivot, a tractor, a planter or a sprayer.

4. Electrical weeding device according to any of claims 1 to 3, characterized by the fact that it comprises:a first electrode located at the tip of an insulated arm connected to the drone; anda second electrode grounded in a remote location.

5. System for applying electrical current into undesired plants, characterized by the fact that it comprises:at least one support structure positioned in the area to be treated, serving as a docking and power supply station;a tethered drone connected to the support structure, configured for autonomous horizontal and vertical movement, with real-time position adjustments based on GPS / GNSS monitoring;at least one vision sensor connected to a processing unit, comprising RGB and RGB-D cameras, LiDAR, and ultrasonic sensors, configured tocapture images of the area to be treated and identify undesired plants through a segmentation and crop row detection process;at least one electrode connected to an electrical converter, configured to apply electrical current by contacting the undesired plant;an electrical converter configured to adjust the parameters of the electrical current for performing the weeding process; anda control module configured to coordinate the movement of the drone, the operation of the vision sensor, and the application of electrical current to the target plants.

6. System according to claim 5, characterized by the fact that it operates through a physical layer comprising at least two of the following components: a drone, vision sensor, actuator, carrier, electrode, and tether.

7. System according to claim 5 or 6, characterized by the fact that it operates through a logic layer comprising at least two of the following components: a processing unit configured to process sensor data, flight control functionality, computational logic for determining drone actions, a database for storing system and application information, and georeferencing of the target plant location.

8. System according to any of claims 5 to 7, characterized by the fact that it operates through an application layer configured to perform at least two of the following actions: flight planning, mission control, operation of electrodes or actuators, and recording flight data.

9. Method for applying electrical current to undesired plants, characterized by the fact that it comprises the steps of:moving a drone tethered to a support structure across an area to be treated, with autonomous navigation based on GPS / GNSS monitoring and onboard flight control;capturing image data of the area with a vision sensor mounted on the drone, comprising RGB and RGB-D cameras, LiDAR, and ultrasonic sensors;processing the image data with a processing unit to identify and locate undesired plants;transforming the identified plant positions from image coordinates into the local reference system through a transformation frame to define precise actuation points;positioning an electrode connected to the drone above the identified undesired plant;moving the electrode toward the undesired plant using a vertical movement mechanism; andapplying electrical current to the undesired plant through the electrode, with the electrical current parameters adjusted by an electrical converter based on the plant’s characteristics.