Power inspection foreign matter removing control method and control system based on vehicle-mounted unmanned aerial vehicle

Abstract

The application discloses a power inspection foreign matter removing control method and control system based on a vehicle-mounted unmanned aerial vehicle, and belongs to the technical fields of unmanned aerial vehicles and power inspection. The method comprises the following steps: after system self-checking, the unmanned aerial vehicle autonomously takes off and flies to a target area; the foreign matter category is identified and the relative distance is calculated through binocular vision; the identification information is sent to the vehicle-mounted platform for manual confirmation; after receiving the confirmation, the unmanned aerial vehicle adaptively adjusts the flight attitude and the spraying parameter according to the foreign matter type, and executes the removing; and after the task is completed, the unmanned aerial vehicle automatically returns and carries out energy supply. The application realizes wide-area maneuvering and automatic supply of the unmanned aerial vehicle through the vehicle-mounted mobile platform, realizes accurate, safe and efficient removing of the foreign matter on the power transmission line by combining a man-machine cooperative decision mechanism and multi-parameter adaptive control based on the foreign matter material and distance, and greatly improves the automation and intelligent level of power inspection.

Description

Technical Field

[0001] This invention relates to the fields of unmanned aerial vehicles (UAVs) and power line inspection technology, specifically to a method and control system for removing foreign objects during power line inspection based on a vehicle-mounted UAV. Background Technology

[0002] High-voltage transmission lines are mostly exposed to the natural environment, making them prone to entanglement with foreign objects such as plastic film, kite strings, and wasp nests, seriously threatening the safe and stable operation of the power grid. Traditional methods of removing foreign objects mainly rely on manual tower climbing or live-line work, which are labor-intensive and pose high safety risks.

[0003] In recent years, with the development of drone technology, technical solutions have emerged that use drones to carry flamethrowers or laser devices for foreign object removal. For example, Chinese invention patent CN117638714A, entitled "A Method and Apparatus for Removing Foreign Objects from High-Altitude Transmission Lines," discloses a method for removing foreign objects from transmission lines using a drone. This method includes: during drone inspection, scanning and identifying foreign object targets in the inspection area using an onboard intelligent camera; controlling the drone to fly towards the foreign object target and acquiring the real-time distance between the drone and the foreign object target; when the real-time distance reaches a first preset distance, adjusting the drone's flight direction based on the foreign object target's position in the captured image; when the real-time distance reaches a second preset distance, adjusting the drone's hovering angle based on the positional relationship between the foreign object target and the ignition positioning point; wherein the second preset distance is less than the first preset distance; based on the hovering angle, controlling the drone to hover at the second preset distance and detecting the type and area of ​​the foreign object target; determining the solenoid valve setting based on the type and area of ​​the foreign object target, and igniting the foreign object target based on the solenoid valve setting. This method can effectively burn away foreign objects on power transmission lines, avoiding high-altitude operations for power inspection personnel and significantly improving safety.

[0004] However, this method also has some problems: The spray parameters are not adjustable: The flamethrower is usually fixed to the bottom of the drone, and its spray distance, angle and flame intensity cannot be adjusted in real time according to the material (such as chemical fiber, plant stem) and distance of the foreign object, resulting in low removal efficiency or the risk of igniting surrounding circuits.

[0005] Insufficient intelligence: Existing systems mostly rely on manual operation by operators, which makes it difficult to identify small foreign objects (such as kite strings) in strong light or at long distances, and it is also difficult for drones to adjust their attitude near high-voltage lines.

[0006] Battery life and logistical bottlenecks: Drones have limited battery life and small fuel capacity, making it difficult to support continuous multi-site inspection needs, and they lack deep integration with vehicle-mounted platforms. Summary of the Invention

[0007] The purpose of this application is to address the shortcomings of the aforementioned background technology and provide a method and control system for removing foreign objects during power line inspections based on vehicle-mounted drones.

[0008] The technical solution of this application is: a power line inspection foreign object removal control method based on vehicle-mounted drones, comprising: After receiving the inspection task instruction, the vehicle-mounted mobile platform and the operation drone will perform a system self-check to determine whether they meet the operation requirements. When the operational requirements are met, the drone is controlled to take off and fly to the target tower area, collecting real-time images during the flight. Once the drone reaches the target area, it performs reasoning analysis on the currently acquired images, identifies the type of foreign object, and calculates the relative distance between the drone and the foreign object. A confirmation request packet containing the type of foreign object, its location information, and relative distance is sent to the human-machine interface module of the vehicle-mounted mobile platform, where the operator confirms the removal. If a confirmation command is received, the drone will automatically adjust its flight attitude and spray parameters according to the type of foreign object, and perform a spraying operation to burn and remove the foreign object. After the cleanup mission is completed, the drone automatically returns to the vehicle-mounted mobile platform for refueling.

[0009] According to the power inspection foreign object removal control method based on vehicle-mounted drone provided in this application, the method for system self-testing of vehicle-mounted mobile platform and operation drone includes: comparing the current power of operation drone with power threshold, comparing the remaining fuel of operation drone with fuel threshold, and if the current power is not lower than the power threshold and the remaining fuel is not lower than the fuel threshold, then the current operation drone is determined to meet the operation requirements; otherwise, it does not meet the operation requirements.

[0010] According to the power line inspection foreign object removal control method based on vehicle-mounted UAV provided in this application, the method for calculating the relative distance between the UAV and the foreign object includes: the UAV acquiring real-time images through a binocular vision module during flight, and calculating the relative distance between the UAV and the foreign object based on the following formula: D = (f × b) / d Where: D is the relative distance between the drone and the foreign object, f is the focal length of the left and right cameras in the binocular vision imaging module, b is the baseline distance between the left and right cameras in the binocular vision imaging module, and d is the pixel offset of the foreign object in the left and right images.

[0011] According to the power line inspection foreign object removal control method based on vehicle-mounted UAV provided in this application, the method of automatically adjusting the flight attitude of the UAV according to the type of foreign object includes: determining the recommended working distance according to the type of foreign object; and comparing the current relative distance with the recommended working distance in real time. If the current distance is greater than the sum of the recommended operating distance and the allowable deviation threshold, control the drone to fly forward and approach the target; If the current distance is less than the difference between the recommended operating distance and the allowable deviation threshold, control the drone to fly backward away from the target; Repeat the above adjustment process until the absolute value of the difference between the current relative distance and the recommended working distance is no greater than the threshold for stopping the adjustment.

[0012] According to the power line inspection foreign object removal control method based on vehicle-mounted UAV provided in this application, the method for adjusting the spray parameters of the UAV includes: adjusting the spray duration according to the following formula: in: T This refers to the duration of the spray. T base Base spray duration; k d This is the distance compensation coefficient; D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object.

[0013] According to the power line inspection foreign object removal control method based on a vehicle-mounted drone provided in this application, the method for adjusting the spray parameters of the drone includes: adjusting the spray flame intensity according to the following formula: in: P The intensity of the ejected flame; P max For maximum safety strength; P base Basic strength; D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object; or The material combustion difficulty coefficient is determined based on the type of foreign object. Adjust the output power of the high-pressure generator and the opening of the fuel supply valve according to the calculated flame intensity.

[0014] According to the foreign object removal control method for power grid inspection based on a vehicle-mounted drone provided in this application, the combustion difficulty coefficient of the material is... or The setting rules are as follows: When the foreign object is identified as plastic film, the following settings are configured: or =1.0; When the foreign object category is identified as a damp wasp nest, set or =1.5.

[0015] According to the power inspection foreign object removal control method based on vehicle-mounted UAV provided in this application, the method of performing the spraying operation includes: continuously monitoring the foreign object status through a binocular vision module during the spraying process; and immediately analyzing whether there is still a foreign object marker box in the current frame image after a single spraying. If a foreign object marker box is present and the current cumulative number of sprays is less than the preset maximum number of cycles threshold, then execute the next round of spraying; If the foreign object marker disappears or the cumulative number of sprays reaches the threshold, the spraying operation will terminate and the device will be suspended.

[0016] According to the power inspection foreign object removal control method based on vehicle-mounted drone provided in this application, the energy replenishment method includes: after the drone lands, the charging mechanism automatically extends and makes physical contact with the drone charging contacts, negotiates the charging power with the vehicle power supply through the battery management system (BMS), and executes a constant current and then constant voltage charging mode. The system detects the pressure inside the drone's fuel tank. If the pressure is lower than the set value, the solenoid valve is automatically opened, and the fuel is injected into the drone's fuel tank using the pressure inside the vehicle's fuel tank.

[0017] This application also relates to a power line inspection foreign object removal control system based on a vehicle-mounted drone, used to implement the above-mentioned method, including: The vehicle-mounted mobile platform includes an integrated management and control platform, an intelligent hangar, and a first communication module; the integrated management and control platform includes a task planning module, a human-machine interaction module, and a data link module; the intelligent hangar includes a helipad, an automatic charging device, and a fuel replenishment device. The operational drone is a multi-rotor structure equipped with a three-axis self-stabilizing gimbal, a binocular vision imaging module, a foreign object identification and processing module, a distance calculation unit, a multi-degree-of-freedom jet device, a flight control system, and a second communication module. The multi-degree-of-freedom spraying device is mounted on the three-axis self-stabilizing gimbal and includes a spray gun, a high-pressure generator, and a fuel cylinder. The spraying direction of the spray gun is independently controlled by the gimbal. The vehicle-mounted mobile platform and the operating drone establish a data link through the first and second communication modules to realize video stream transmission, control command issuance and status data feedback, and jointly complete the entire process of operation from automatic resupply, autonomous inspection, human-machine collaborative decision-making to adaptive clearing.

[0018] The advantages of this application are: 1. Vehicle-mounted mobility and automatic resupply enable wide-area sustainable operations. The system combines operational drones with a vehicle-mounted mobile platform integrated with an intelligent hangar. The vehicle serves as a mobile base, significantly expanding the inspection range of a single mission. More importantly, after completing its mission, the drone can automatically return to base, where the hangar automatically performs precise charging docking and refueling, achieving true unmanned energy replenishment. This mobile base station and automatic refueling design enables the system to operate continuously for extended periods, over long distances, and at high frequencies, eliminating reliance on frequent manual refueling.

[0019] 2. Human-machine collaborative decision-making and high-precision perception ensure safe and reliable operation. The solution does not aim for fully automated removal, but rather establishes an efficient human-machine collaboration mechanism: the drone autonomously detects foreign objects and calculates their precise location using binocular vision and recognition modules, then pushes the information to the vehicle-mounted platform for final confirmation by experienced operators. This model of machine recognition and human confirmation leverages the advantages of high-precision perception (high spatiotemporal resolution) and rapid response from machines while introducing human experience as a safety redundancy, effectively preventing misjudgments and misoperations, and ensuring absolute safety during operations in complex power environments.

[0020] 3. Adaptive parameter adjustment and closed-loop control improve cleaning accuracy. The drones demonstrated a high degree of intelligence and adaptability in the cleanup operation. Based on the identified type of foreign object (such as plastic film or wasp nests), they could automatically determine the appropriate recommended operating distance and combustion difficulty level, and dynamically adjust their flight attitude through a closed-loop feedback algorithm. Simultaneously, the system could comprehensively consider distance deviation and material characteristics to calculate and adjust the spray duration and flame intensity in real time and with precision, achieving refined control of the spray energy and ensuring optimal cleanup results.

[0021] 4. Fully automated closed-loop process, reducing manual intervention and operational difficulty. Upon receiving the inspection mission instruction, the system enters a highly automated closed-loop process: automatically performing system self-checks, autonomous takeoff and cruise, autonomous distance measurement, human-machine collaborative confirmation, adaptive jet cleaning, and automatically returning to base for resupply after mission completion, preparing for the next mission. Except for the one-click confirmation step, the entire process requires no manual intervention, greatly simplifying the operation, reducing the skill requirements for operators, and achieving truly unmanned operation.

[0022] 5. The system design enhances environmental adaptability and ensures performance under complex operating conditions. The system's key component design fully considers the impact of complex environments. The operational UAV is equipped with a three-axis self-stabilizing gimbal, ensuring the stability of visual imaging and the multi-degree-of-freedom jet propulsion system even under adverse weather conditions such as strong winds. Simultaneously, the use of vision-based binocular ranging technology enables the UAV to maintain high-precision spatial positioning capabilities even in mountainous and densely forested areas where GPS signals are weak or lost. These design features significantly enhance the system's adaptability to different weather conditions and geographical environments, ensuring reliable performance under various complex operating conditions. Attached Figure Description

[0023] Figure 1 This application presents a schematic diagram of the foreign object removal process for power line inspection based on a vehicle-mounted drone. Detailed Implementation

[0024] The embodiments of this application are described in detail below, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0025] In the description of this application, it should be understood that the terms "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0026] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0027] The present application will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0028] This application relates to a power line inspection foreign object removal system and method based on a vehicle-mounted drone, aiming to solve the problems of low efficiency and high risk in traditional manual inspections, and the short battery life and insufficient intelligence of existing drone solutions. The vehicle-mounted platform of this application integrates a comprehensive management and control system and an intelligent hangar, possessing task planning, human-machine interaction, and automatic energy replenishment functions; the drone is equipped with binocular vision, a foreign object recognition module, and a multi-degree-of-freedom spray device to achieve precise perception and removal. This application achieves wide-area mobility and automatic replenishment, a human-machine collaborative decision-making mechanism, balancing efficiency and safety; multi-parameter adaptive control based on foreign object material and distance improves removal accuracy; visual closed-loop monitoring and fault-tolerant mechanisms ensure task reliability. This solution significantly improves the automation, intelligence, and environmental adaptability of power line inspections.

[0029] Specifically, such as Figure 1 As shown, the core of this application's method for controlling the removal of foreign objects during power line inspections based on a vehicle-mounted drone lies in constructing a complete closed-loop operation process. This method includes: S1. Upon receiving the inspection task instruction, the system performs a self-check on the vehicle-mounted mobile platform and the operating drone to ensure that all equipment statuses meet the requirements for task execution. Once the self-check is passed, the system controls the operating drone to automatically take off from the vehicle-mounted intelligent hangar and fly to the target tower area along a preset route.

[0030] S2. During flight, the drone continuously collects and transmits real-time images through its onboard vision module.

[0031] S3. When the drone arrives at the target area and detects a suspected foreign object, it performs local or edge-level reasoning analysis on the currently acquired images to identify the specific type of foreign object (such as plastic film, kite, wasp nest, etc.) and calculates the precise relative distance between the drone and the foreign object in real time.

[0032] S4. The drone will send a confirmation request packet containing the type of foreign object, geographical location information, and calculated relative distance to the human-machine interaction module of the vehicle-mounted mobile platform via a wireless data link. Experienced operators will then conduct the final manual confirmation and issue a removal command.

[0033] S5. If a confirmation command is received, the drone will automatically adjust its flight attitude and the parameters of the spray device according to the type of foreign object identified, and then perform a precise spraying operation to burn and remove the foreign object.

[0034] S6. After the clearing task is completed, the drone will automatically leave the pole area and return to the vehicle-mounted mobile platform according to the preset route. Finally, it will automatically recharge its energy and wait for the next mission command.

[0035] Principle description: The principle behind this embodiment lies in the organic integration of six stages: system self-check, autonomous flight, intelligent identification, manual confirmation, precise removal, and automatic refueling, thus constructing a closed-loop operation process that is either unmanned or requires minimal human intervention. System self-check is a prerequisite for safe flight, ensuring the drone has sufficient energy to complete the mission. Autonomous flight and intelligent identification utilize the drone's high maneuverability and onboard AI computing power, extending the inspectors' eyes and ears to high altitudes and distant locations. The manual confirmation stage provides crucial safety redundancy, using human experience to avoid potential misjudgments by the machine algorithm. Precise removal relies on understanding the physical characteristics of different foreign objects and pre-setting different operational parameters. Automatic refueling ensures the system's continuous operational capability, eliminating the need for tedious on-site battery replacement and refueling operations.

[0036] This application achieves full-process automation and intelligence. Inspection personnel only need to issue instructions and make key confirmations from inside the vehicle to complete complex foreign object removal tasks, greatly reducing labor intensity and operational difficulty, and minimizing personnel input. The closed-loop design ensures that each deployment is a complete and effective operation, improving the effective utilization rate of the inspection area. Furthermore, through human-machine collaboration, operational safety is ensured while maintaining operational efficiency.

[0037] In some embodiments of this application, the self-testing method described above has been optimized. Specifically, the method includes: during the self-testing process, the system compares the current battery level of the drone with a preset battery threshold (in this embodiment, the battery threshold is 80% of the total battery level; in actual applications, it is not limited to this value and can be set according to requirements), and compares the remaining fuel level of the drone with a preset fuel threshold (in this embodiment, the fuel threshold is 400ml; in actual applications, it is not limited to this value and can be set according to requirements; or a pressure comparison can be performed, i.e., setting a pressure threshold for the fuel tank). Only when the current battery level is not lower than the battery threshold and the remaining fuel level is not lower than the fuel threshold, does the system determine that the drone meets the operational requirements and allow it to take off and perform the task; otherwise, the system determines that the operational requirements are not met and displays the specific non-compliance item on the human-machine interface (e.g., "Battery level is low, please wait to charge" or "Fuel level is low, please refuel"), while preventing the drone from taking off.

[0038] Upon receiving the mission command, the vehicle-mounted integrated control platform sends a "request status data" command to the drone via its communication module. The drone's flight control system responds to this command, immediately reading the current state of charge (SOC) of the battery pack through its internal battery management system (BMS) and the remaining fuel level via pressure or level sensors on the fuel tank. The drone then packages this real-time data into a status frame and transmits it back to the vehicle-mounted platform via a second communication module. The vehicle-mounted platform's data processing unit then parses the battery level value. E real and fuel value F real and compare it with a fixed threshold stored in the local database (e.g. E threshold =4800mAh, corresponding to 80% of the total battery capacity of 6000mAh; F threshold =600ml) to be compared. The comparison logic is "logical AND", that is E real ≥ E threshold and F real ≥ F threshold If the result is true, the platform sends a "takeoff permitted" command to the drone, and the mission continues. If the result is false, the platform aborts the mission and highlights "Self-check failed" and the specific reason in the status bar of the human-machine interface, such as "Battery power 45%, needs charging" or "Fuel remaining 50ml, needs replenishing".

[0039] The principle behind this embodiment is to simplify the complex assessment of operational capabilities into a comparison of thresholds for two key physical quantities—electrical energy and fuel energy. Electricity and fuel are the fundamental energy sources for the UAV to complete its flight and propulsion missions. By setting conservative thresholds, it can be ensured that the UAV not only has enough energy to fly to the target point, but also to complete a series of energy-intensive operations such as hovering, adjustments, and propulsion at the target point, and has sufficient margin for a safe return. This "dual threshold" comparison method is a simple, reliable, and efficient logical judgment method that avoids mission failure or flight accidents due to insufficient energy.

[0040] This embodiment provides clear and reliable criteria for task initiation, fundamentally ensuring the feasibility and success rate of the task. The specific power and fuel threshold settings demonstrate the solution's emphasis on operational safety and reliability. Simultaneously, the automatic detection and prompting functions eliminate the need for operators to rely on experience for estimation, further simplifying operations and improving the adaptability and ease of use of the vehicle-mounted platform.

[0041] In other embodiments of this application, the ranging method described above is optimized. Specifically, the method employs binocular vision technology. During flight, the UAV simultaneously acquires two images, one on the left and one on the right, containing the foreign object, using its onboard binocular vision imaging module. Subsequently, the distance calculation unit calculates the relative distance between the UAV and the foreign object based on the following formula: D=(f×b) / d Where D is the calculated relative distance between the drone and the foreign object; f is the focal length of the left and right cameras in the binocular vision imaging module (in pixels); b is the baseline distance between the left and right cameras in the binocular vision imaging module (in meters); and d is the pixel offset (i.e., parallax, in pixels) of the same foreign object in the left and right images calculated by the stereo matching algorithm.

[0042] Once the airborne foreign object detection and processing module identifies a foreign object in a frame and generates a foreign object marker box, it immediately triggers the distance calculation unit. The distance calculation unit first retrieves the left and right original images with the closest timestamp to the current frame from the binocular vision module's cache. Then, based on the center point or specific feature point of the foreign object marker box in the left image, it searches for corresponding points in the right image using epipolar constraints and block matching algorithms, calculating the pixel coordinate difference of that point, i.e., the disparity value d. Next, the calculation unit reads the pre-calibrated camera focal length f and baseline distance b from the system parameter area. Finally, substituting f, b, and d into the above formula, it calculates the precise depth distance D. This distance value D is then packaged into a confirmation request packet along with the foreign object category and location information.

[0043] This embodiment is based on the principle of triangulation. Binocular vision systems mimic the way the human eye perceives depth. When two parallel cameras, with a known distance (b), photograph the same object, the object's image position in the left and right images will have a horizontal offset (parallax d). The magnitude of this parallax is inversely proportional to the distance from the object to the camera. With a fixed camera focal length f, the distance to the object can be calculated using geometric formulas by accurately measuring the parallax d. Because this method is based on pixel-level calculations, it can provide very high-precision distance information.

[0044] This embodiment acquires image information in real time through a binocular vision imaging module, and runs a target detection network through an onboard AI acceleration unit to identify plastic film, kite string, or hornet's nest in the image, generating a bounding box containing the outline of the foreign object. Essentially, it compares the acquired image information with images of plastic film, kite string, or hornet's nest in a database. If an object resembling plastic film, kite string, or hornet's nest appears in the acquired image information, and the similarity between the object and the image exceeds a set similarity threshold, the type of foreign object can be determined. Then, through image processing, the outline of the foreign object is identified, and a bounding box for the foreign object is obtained.

[0045] This embodiment provides a high-precision, high-frame-rate distance measurement method. Compared to GPS or ultrasonic ranging, binocular vision ranging offers extremely high spatiotemporal resolution, providing centimeter-level distance information in real-time as the UAV approaches a tower, forming the basis for subsequent precise adjustments to flight attitude and jet parameters. Furthermore, this method relies entirely on airborne vision, without depending on external satellite signals or ground base stations, allowing it to operate stably even in mountainous areas with weak GPS signals or in environments with strong electromagnetic interference, significantly improving the system's environmental adaptability and reliability. Data homogeneity (both recognition and ranging are vision-based) also reduces system errors.

[0046] In a further embodiment of this application, the method for automatically adjusting the flight attitude of the above-mentioned operational drone based on the type of foreign object is optimized. The method includes: first, querying a recommended operational distance corresponding to the identified foreign object category from an onboard database. D ref (The recommended working distance is obtained by looking up a table based on the object category; for example, the recommended working distance for plastic film is 4m, for kite string it is 5m, and for wasp nests it is 3m.) During the spraying preparation phase, the flight control system acquires the current relative distance D in real time. Then, the system compares the current distance D with the recommended working distance. D ref The comparison is performed, and an allowable deviation threshold Δ and a stop adjustment threshold ε are introduced (typically...). e<D The specific adjustment logic is as follows: like D>(D ref +D) If the distance between the drone and the target is too far, the drone will be controlled to fly forward and gradually approach the target. like D<(D ref -D) If the drone is too close to the target, it will be controlled to fly backward and gradually move away from the target. like (D ref -Δ)≤D≤(D ref +D) If the condition is not met, the drone is considered to be within the coarse adjustment range, but may not necessarily meet the fine adjustment requirements.

[0047] Subsequently, the drone entered the fine-tuning phase, continuously comparing D with... D ref ,if only |DD ref |>e Then continue adjusting forward or backward in small steps, repeating the above comparison and adjustment process until... |D-Dref|≤εAt this point, it is determined that the drone has accurately reached the optimal operating position and stops attitude adjustment.

[0048] Assuming the foreign object identification result is a plastic film, the recommended operating distance corresponding to the airborne database is... Dref The distance is 4 meters. The system's allowable deviation threshold Δ is set to 0.5 meters, and the stop adjustment threshold ε is set to 0.1 meters. After initially confirming its position, the UAV measures the current distance D to be 4.8 meters. The flight control system first calculates and compares: 4.8 > (4.0 + 0.5 = 4.5), the condition is met, so it sends a "forward" command to the flight controller at a speed of 0.2 m / s. During forward movement, the system continuously measures the distance. When the distance becomes 4.4 meters, 4.4 < 4.5 and 4.4 > 3.5, entering the coarse adjustment range, the UAV stops moving forward and hovers. At this point, it enters the fine adjustment stage. The system finds that |4.4 - 4.0| = 0.4 > 0.1, so it again controls the UAV to make fine adjustments forward at an ultra-low speed of 0.05 m / s. When the distance becomes 4.08 meters, |4.08 - 4.0| = 0.08 ≤ 0.1, the system determines that the position is accurate and issues a "stop adjustment" command. The UAV hovers stably at this position, preparing for jet propulsion.

[0049] The core principle of this embodiment is the introduction of a two-stage closed-loop feedback control strategy. The recommended operating distance is the control target, and the current relative distance is the feedback quantity. First, a rapid coarse adjustment is performed using a relatively wide allowable deviation threshold to quickly bring the UAV near the target area, thereby improving response speed. Then, a more stringent stop adjustment threshold is used for fine-tuning to ensure the final positioning accuracy. This coarse-to-fine control method balances adjustment efficiency and final operating accuracy, enabling the UAV to stably stop at the most suitable and precise position for removing foreign objects.

[0050] This embodiment enables the drone to intelligently adjust and precisely locate itself based on the type of foreign object. Through the setting of two threshold levels, a perfect combination of rapid response and high-precision positioning is achieved. This ensures that regardless of where the drone initially lands, it will ultimately remain stable within the optimal operating window. This not only guarantees the best effect of subsequent spraying and removal (appropriate distance) but also minimizes damage to electrical equipment or the drone itself due to improper distance, significantly improving the safety and intelligence of the operation.

[0051] In a preferred embodiment of this application, the method for adjusting the spray parameters is optimized. Specifically, this embodiment includes adjusting the spray duration and adjusting the flame intensity. Depending on the type of foreign object, the drone can use these two methods individually or in combination.

[0052] Adjust the injection duration according to the following formula: in: TThis refers to the duration of the spray. T base Base spray duration (a preset baseline value, such as 3 seconds); k d This is the distance compensation coefficient (a positive number used to adjust the degree of influence of distance deviation on duration, such as 0.5s). D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object.

[0053] Adjust the intensity of the jet flame according to the following formula: in: P The intensity of the ejected flame; P max The maximum safe strength (a set value, for example, it can be set to the strength corresponding to the maximum opening of the injection valve); P base The base strength (a set value, for example, it can be set to the strength corresponding to 80% opening of the injection valve); D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object; or The combustion difficulty coefficient of the material is determined based on the type of foreign object.

[0054] This embodiment addresses the combustion difficulty coefficient of the material. or The setting rules are illustrated as follows: when the foreign object is identified as a plastic film, its material is relatively flammable, so η is set to 1.0; when the foreign object is identified as a damp wasp nest, its material has a high water content and is difficult to ignite and burn, so η is set to 1.5, which means that 1.5 times the energy of the base intensity is required to effectively remove it.

[0055] Assuming the drone has been positioned optimally, the current distance is D = 4.08 meters, and the foreign object is a damp wasp nest, D ref =4.0 meters.

[0056] First, the system calculates the injection duration. Let... T base =3 seconds, k d =0.5 seconds / meter. Therefore, T = 3 + 0.5 * (4.08 - 4.0) = 3 + 0.04 = 3.04 seconds. Due to the slightly longer distance, the spraying time is slightly extended.

[0057] Simultaneously, the system calculates the flame intensity. Let... P max =100%,P base =80%, or =1.5. Therefore, P = min(100%, 80% * (4.0 / 4.08) * 1.5) = min(100%, 80% * 0.96 * 1.5) = min(100%, 115.2%) = 100%. The calculated theoretical strength exceeds the maximum safe strength, therefore the final output is limited to... P max =100%. The injection control system then adjusts the output power of the high-pressure generator to the maximum and fully opens the fuel valve, preparing to execute a powerful injection lasting 3.04 seconds.

[0058] The principle of this embodiment is based on a refined modeling of the physical process. Small deviations in distance can affect the heat reaching the target object; this effect can be compensated for by adjusting the spray time. This embodiment not only considers the influence of distance but also introduces a material combustion difficulty coefficient η. This coefficient reflects the impact of the physical and chemical properties of different foreign objects (such as ignition point, calorific value, and moisture content) on the ease of removal. By integrating the two key factors of distance and material into a single formula, and using… P max By implementing safety limiting, precise and safe control of the jet energy is achieved.

[0059] This embodiment achieves intelligent, multi-dimensional adaptive adjustment of spray parameters. Through combined control of duration and intensity, it can output the most suitable energy pack for foreign objects of different distances and materials. In particular, the introduction of material difficulty coefficients allows the drone to employ drastically different removal strategies for plastic film and damp wasp nests, much like an experienced operator. This not only significantly improves removal effectiveness and success rate but also... P max It ensures operational safety and saves valuable fuel by avoiding over-injection, demonstrating a high level of intelligence and precision in operation.

[0060] In some embodiments of this application, the method for performing the spraying operation described above has been optimized. The method includes: throughout the entire spraying process, the UAV continuously acquires real-time images through a binocular vision module and sends them to a foreign object identification and processing module for monitoring. After each single spray, the system immediately pauses subsequent actions and analyzes the high-definition image of the current frame to determine whether a foreign object marker box corresponding to the target foreign object still exists in the image.

[0061] If the analysis results show that foreign object marker boxes are still present in the image, it indicates that the foreign object has not been completely removed. At this time, the system checks an internal counter—the current cumulative number of sprays. If this number is less than a preset maximum cycle count threshold (e.g., 3 times), the system automatically prepares and executes the next round of spraying (including readjusting the attitude and parameters).

[0062] If the analysis results show that the foreign object marker box has disappeared, indicating that the foreign object has been successfully removed, the system will immediately terminate the spraying operation and switch to hovering mode, preparing to return to base. Similarly, if the foreign object marker box has not disappeared, but the cumulative number of sprays has reached the preset maximum cycle count threshold, the system will also forcibly terminate the spraying operation for safety reasons, record a warning message of removal failure in the log, and then hover or return to base.

[0063] The drone begins its first spray. After the spray, the flight control system issues a command to pause all actions. The foreign object detection module immediately captures and analyzes the image at this moment. The detection algorithm searches the image for foreign objects that match the previously detected features. If a foreign object marker box is found (for example, a wasp nest is still there, but the surface is somewhat charred), the counter changes from 0 to 1. The system compares 1 < 3 (threshold), the condition is met, and a second spray is prepared. After the second spray, the image is analyzed again. If the foreign object marker box disappears, the system determines that the removal was successful, terminates the spray, sends a status message indicating that the foreign object has been removed to the vehicle platform, and then initiates the return-to-home procedure. If the foreign object is still present after the second spray, a third spray is performed. After that, regardless of whether the foreign object has disappeared, the system will forcibly terminate the spray and record the event when the counter reaches 3.

[0064] The principle of this embodiment is to introduce a closed-loop feedback of perception and action during the removal process. Visual perception not only serves as a reconnaissance method before removal but also becomes a tool for evaluating the effectiveness of the removal process. Through a cycle of spraying, observing, judging, and re-spraying, real-time verification and repeated reinforcement of the removal effect are achieved. At the same time, introducing a maximum number of cycles as a termination condition is to avoid infinite loop operations caused by the special properties of foreign objects (such as misidentification of metal parts) or system failures. This is a fault-tolerant mechanism to ensure system safety and efficient use of resources.

[0065] This embodiment significantly improves the reliability and thoroughness of the removal task. Through real-time feedback, the system can autonomously determine whether the task is complete, ensuring successful removal of foreign objects in a single attempt and avoiding repeated failures. The setting of the maximum number of cycles serves as a crucial safety measure, preventing ineffective operations and resource waste due to identification errors or other anomalies, demonstrating the system's intelligence and robustness. The entire process requires no manual intervention to assess the removal effect, further achieving unmanned operation.

[0066] In other embodiments of this application, the energy replenishment method is optimized. Specifically, regarding automatic charging: after the drone precisely lands on the helipad of the vehicle-mounted mobile platform, the charging mechanism in the hangar automatically extends, and its charging contacts, guided by a mechanical guide device, reliably make physical contact with the charging contacts on the bottom of the drone. After establishing a connection, the battery management system (BMS) on the drone and the vehicle-mounted power management system negotiate via a communication protocol to jointly determine the optimal charging power and charging mode. Subsequently, the charging process begins, strictly following a constant current followed by a constant voltage charging pattern until the battery is fully charged.

[0067] Regarding automatic refueling: The hangar's refueling system activates simultaneously with or after charging. First, a pressure sensor detects the current gas pressure in the drone's fuel tank. If the detected pressure is below a preset full-capacity pressure threshold (e.g., below 90% of full capacity), the system determines that refueling is needed. At this point, the hangar control system automatically opens the solenoid valve connecting the vehicle-mounted high-pressure fuel tank and the drone's fuel tank. Utilizing the pressure difference between the vehicle-mounted fuel tank and the drone's fuel tank, fuel is automatically pressurized and injected into the drone's fuel tank. When the pressure in the fuel tank reaches the preset full-capacity threshold, the pressure sensor sends a feedback signal, and the control system immediately closes the solenoid valve, completing the refueling process.

[0068] The drone returned and landed on the helipad, where the hangar's positioning pins assisted in precise docking. Subsequently, a charging arm with a universal joint rose, its two charging probes inserting into the charging port on the drone's bottom. The BMS detected the external power connection and communicated with the onboard charger via data cable, reporting the battery's current voltage, temperature, and required current. Based on the BMS's requirements, the onboard charger began charging at a constant current of 0.5C. When the battery voltage reached the charging limit voltage, it automatically switched to constant voltage charging, with the current gradually decreasing until it dropped to 0.05C, at which point charging was complete, and the charging arm automatically retracted. Simultaneously, the fuel replenishment unit's detection head connected to the drone's fuel tank filling port, detecting an internal pressure of 0.5MPa, lower than the set full pressure of 1.0MPa. The control system opened the solenoid valve, and fuel at 1.5MPa from the onboard high-pressure fuel tank rapidly flowed into the drone's fuel tank. When the internal pressure rose to 1.0MPa, the pressure switch activated, the solenoid valve closed, and refueling was complete.

[0069] The principle of this embodiment is to achieve automated and safe replenishment of two different forms of energy. For electrical energy, a physical contact connection is used, and a BMS intelligent negotiation charging strategy ensures the safety of the charging process and battery life. Constant current followed by constant voltage is the optimal charging mode for lithium batteries. For fuel, a pressure differential injection method is used, which is a simple and reliable non-powered refueling method that avoids the use of an additional pump, reducing system complexity and failure rate. Pressure monitoring ensures precise control of the refueling amount and prevents overcharging.

[0070] This embodiment truly achieves a closed-loop automatic refueling system, enabling the entire system to operate autonomously for extended periods and at high frequencies without human intervention. The intelligent charging strategy ensures the health and lifespan of the battery, a core component. Automatic and precise refueling ensures the drone is always fully fueled for every sortie. This not only significantly reduces the logistical burden and operational risks for operators (no contact with high-pressure fuel is required), but also gives the system true unmanned operation and continuous cleaning capabilities, providing the most fundamental energy guarantee for a substantial increase in the inspection range.

[0071] This application also provides a power line inspection foreign object removal control system based on a vehicle-mounted drone, which is used to implement the above-mentioned method. The system mainly consists of two parts: a vehicle-mounted mobile platform and an operating drone.

[0072] The vehicle-mounted mobile platform functions as a mobile command center and logistics base. Internally, it houses an integrated control platform, an intelligent hangar, and a primary communication module. The integrated control platform, the brain of the entire system, further includes a mission planning module (for generating flight paths), a human-machine interaction module (for information display and receiving operator instructions), and a data link module (responsible for data packaging and unpacking). The intelligent hangar serves as the physical support unit, equipped with a helipad for UAV takeoff and landing, an automatic charging device for automatic charging, and a refueling device for refueling.

[0073] The described unmanned aerial vehicle (UAV) is a multi-rotor aircraft, serving as the system's vision and execution modules. It is equipped with a three-axis self-stabilizing gimbal for maintaining visual and jet stability, a binocular vision imaging module for image acquisition, a foreign object identification and processing module for detecting foreign objects, a distance calculation unit for distance calculation, a multi-degree-of-freedom jetting device for performing clearance operations, a flight control system for controlling flight, and a second communication module for communicating with the ground. The multi-degree-of-freedom jetting device, mounted on the three-axis self-stabilizing gimbal, includes a spray gun, a high-pressure generator, and a fuel tank. The jetting direction is independently controlled by the gimbal, achieving decoupling between flight attitude and aiming direction.

[0074] The vehicle-mounted mobile platform and the operating drone establish a high-speed, reliable two-way data link through the first and second communication modules. This link enables downlink transmission of high-definition video streams, uplink transmission of control commands, and real-time transmission of drone status data. All these modules work together to complete a closed-loop operation encompassing automatic resupply, autonomous inspection, human-machine collaborative decision-making, and adaptive clearance.

[0075] Once the system starts, the operator sets the inspection task on the human-machine interface module of the vehicle-mounted integrated control platform. The task planning module generates a flight path and uploads the task to the drone via the first communication module. After takeoff, the video stream captured by the drone's binocular vision module is transmitted back in real time via the second communication module and displayed on the human-machine interface module. When the drone's foreign object detection module detects a foreign object, it transmits the result along with the ranging value calculated by the distance calculation unit back to the drone, and the human-machine interface module pops up a confirmation request. After operator confirmation, the command is sent to the drone, and the flight control system controls the drone to fly stably. Simultaneously, the multi-degree-of-freedom jet device, under the independent control of the gimbal, precisely targets the foreign object and removes it. After removal, the drone automatically returns to the helipad of the vehicle-mounted mobile platform. The hangar's automatic charging and refueling devices then activate to replenish the drone's energy, preparing it for the next mission.

[0076] This system organically integrates four functional modules: perception, decision-making, execution, and support. The vehicle-mounted mobile platform solves the problems of mobility and resupply, enabling the system to operate over a wide area continuously. The operational drone solves the problems of observation and execution; its highly integrated vision and spraying system ensures accurate target detection and elimination. In particular, mounting the spraying device on an independently controlled gimbal allows the drone to independently control the direction of the spray gun while adjusting its own position, greatly improving aiming flexibility and accuracy. The high-speed data link, like a nervous system, connects all modules into a collaborative whole.

[0077] The foreign object removal control method for power line inspection based on vehicle-mounted drones in this application specifically includes the following steps: Task reception and system self-check: The vehicle-mounted integrated management and control platform receives inspection task instructions.

[0078] The platform wakes up and queries the status of the operating drone through the first communication module.

[0079] The drone's flight control system responds to queries and displays battery level. E real and fuel balance F real Return.

[0080] The platform will E realWith preset power threshold E threshold , F real Compare with the preset fuel threshold Fth.

[0081] like E real ≥ E threshold and F real ≥ F threshold If the task is deemed to meet the requirements, proceed to the next step; otherwise, display the specific failed item on the human-computer interaction module and terminate the task.

[0082] Autonomous takeoff and cruise identification: The platform sends takeoff and flight path instructions to the drone.

[0083] The drone takes off from the vehicle-mounted mobile platform and flies to the target tower area along a preset route.

[0084] During flight, the binocular vision imaging module on the drone continuously acquires images, while the foreign object identification and processing module performs real-time reasoning and analysis on the images to search for foreign objects.

[0085] Target arrival, identification, and ranging: When the drone arrives at the target area, the foreign object identification and processing module detects and marks the foreign object in the image.

[0086] The trigger distance calculation unit uses the left and right images synchronously acquired by the binocular vision module to calculate the disparity d of the foreign object in the left and right images through a stereo matching algorithm.

[0087] The distance calculation unit reads the pre-calibrated camera focal length f and baseline distance b, substitutes them into the formula D=(f×b) / d, and calculates the precise relative distance D between the drone and the object.

[0088] Human-machine collaboration confirmation: The drone packages the object type, GPS location information, and calculated relative distance D into a confirmation request packet, which is then sent to the vehicle-mounted mobile platform via a data link.

[0089] The platform's human-computer interaction module displays the above information and real-time video to the operator through a graphical interface and voice prompts, requesting confirmation to clear the message.

[0090] After observing and assessing the situation, the operator issues a confirmation or rejection command via the interactive module. If a confirmation command is received, the process continues; otherwise, the drone returns to base as instructed.

[0091] Adaptive flight attitude adjustment: The drone's flight control system retrieves the corresponding recommended operating distance from the onboard database based on the type of foreign object. D ref The allowable deviation threshold Δ and the stop adjustment threshold δ.

[0092] Real-time comparison of current distance D and D ref : Perform a closed-loop adjustment from coarse to fine, until... |DD ref |≤ ε ensures that the drone hovers stably in the optimal operating position.

[0093] Adaptive injection parameter calculation: The injection control system obtains the basic injection duration based on the type of foreign object. T base Distance compensation coefficient k d Foundation strength P base Maximum safety strength P max and the difficulty of burning materials or (such as plastic film) n=1.0 Damp wasp nest n=1.5 ).

[0094] Based on the current precise distance D and recommended distance D ref Calculate the actual spraying time .

[0095] Calculate the actual flame intensity .

[0096] Based on the calculated P value, the output power of the high-voltage generator and the opening of the fuel valve are adjusted via PWM signals.

[0097] Precise removal and effect monitoring: The drone initiates its spraying operation. Under the control of the gimbal, the multi-degree-of-freedom spraying device precisely targets the foreign object and removes it according to the calculated time (T) and intensity (P).

[0098] During the spraying process and after each spraying, the binocular vision module and the foreign object identification and processing module continuously monitor the status of the foreign object.

[0099] After a single spray, analyze the current frame image: If the foreign object marker box disappears: the removal is considered successful, and the spraying process ends.

[0100] If the foreign object marker box remains: Check the cumulative number of sprays n. If n < the preset maximum number of cycles (e.g., 3 times), increase n and return to the above steps, readjusting the attitude and parameters according to the latest status to prepare for the next round of spraying; if n ≥ the maximum number of cycles, the removal is deemed a failure, the spraying is forcibly terminated, and the log is recorded.

[0101] Automatic return and energy replenishment: After the mission is completed (either clearing successfully or recording failed), the flight control system generates a return flight path, controls the drone to fly back above the vehicle-mounted mobile platform, and lands vertically on the tarmac.

[0102] After landing, the charging probe of the automatic charging device reliably connects to the drone's charging contacts.

[0103] The drone's BMS negotiates with the vehicle's power supply to initiate a smart charging process that first uses constant current and then constant voltage until the battery is fully charged.

[0104] Meanwhile, the fuel replenishment system monitors the pressure of the drone's fuel tank. If the pressure is below the threshold, the solenoid valve automatically opens, using the pressure difference from the vehicle's fuel tank to inject fuel into the drone's fuel tank until the pressure reaches the full threshold.

[0105] After resupply is completed, the drone enters standby mode, awaiting the next mission command.

[0106] This application presents a highly intelligent, autonomous, safe, and reliable closed-loop system for removing foreign objects during power line inspections. From energy self-checking to ensure safe takeoff, to high-precision binocular vision perception and identification; from human-machine collaboration to ensure accurate decision-making, to adaptive position and energy control for precise removal; from real-time performance monitoring to ensure complete mission completion, to automatic energy replenishment for continuous operation—every step demonstrates an extremely high level of automation and intelligence. The entire process significantly reduces manual intervention and operational difficulty, improves operational efficiency and safety, expands the effective operational range of a single shift, and can adapt to complex geographical and meteorological environments, representing a major advancement in power line operation and maintenance technology.

[0107] The foregoing has shown and described the basic principles, main features, and advantages of this application. Those skilled in the art should understand that this application is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this application. Various changes and modifications can be made to this application without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of this application as claimed. The scope of protection of this application is defined by the appended claims and their equivalents.

Claims

1. A method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV), characterized in that: include: After receiving the inspection task instruction, the vehicle-mounted mobile platform and the operation drone will perform a system self-check to determine whether they meet the operation requirements. When the operational requirements are met, the drone is controlled to take off and fly to the target tower area, collecting real-time images during the flight. Once the drone reaches the target area, it performs reasoning analysis on the currently acquired images, identifies the type of foreign object, and calculates the relative distance between the drone and the foreign object. A confirmation request packet containing the type of foreign object, its location information, and relative distance is sent to the human-machine interface module of the vehicle-mounted mobile platform, where the operator confirms the removal. If a confirmation command is received, the drone will automatically adjust its flight attitude and spray parameters according to the type of foreign object, and perform a spraying operation to burn and remove the foreign object. After the cleanup mission is completed, the drone automatically returns to the vehicle-mounted mobile platform for refueling.

2. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The method for performing system self-checks on the vehicle-mounted mobile platform and the operation drone includes: comparing the current battery level of the operation drone with a battery threshold, and comparing the remaining fuel level of the operation drone with a fuel threshold. If the current battery level is not lower than the battery threshold and the remaining fuel level is not lower than the fuel threshold, then the operation drone is determined to meet the operation requirements; otherwise, it does not meet the operation requirements.

3. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The method for calculating the relative distance between the drone and the foreign object includes: the drone acquiring real-time images through a binocular vision module during flight, and calculating the relative distance between the drone and the foreign object based on the following formula: D = (f × b) / d Where: D is the relative distance between the drone and the foreign object, f is the focal length of the left and right cameras in the binocular vision imaging module, b is the baseline distance between the left and right cameras in the binocular vision imaging module, and d is the pixel offset of the foreign object in the left and right images.

4. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 3, characterized in that: The method for the drone to automatically adjust its flight attitude according to the type of foreign object includes: determining a recommended operating distance based on the type of foreign object; and comparing the current relative distance with the recommended operating distance in real time. If the current distance is greater than the sum of the recommended operating distance and the allowable deviation threshold, control the drone to fly forward and approach the target; If the current distance is less than the difference between the recommended operating distance and the allowable deviation threshold, control the drone to fly backward away from the target; Repeat the above adjustment process until the absolute value of the difference between the current relative distance and the recommended working distance is no greater than the threshold for stopping the adjustment.

5. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 4, characterized in that: The method for adjusting the jetting parameters of the operational drone includes adjusting the jetting duration according to the following formula: in: T This refers to the duration of the spray. T base Base spray duration; k d This is the distance compensation coefficient; D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object.

6. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 4, characterized in that: The method for adjusting the jet parameters of the operational drone includes adjusting the jet flame intensity according to the following formula: in: P The intensity of the ejected flame; P max For maximum safety strength; P base Basic strength; D The relative distance between the drone and the foreign object; D ref Recommended operating distance determined based on the type of foreign object; η The material combustion difficulty coefficient is determined based on the type of foreign object. Adjust the output power of the high-pressure generator and the opening of the fuel supply valve according to the calculated flame intensity.

7. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 6, characterized in that: The material's fire resistance coefficient η The setting rules are as follows: When the foreign object is identified as plastic film, the following settings are configured: η =1.0; When the foreign object category is identified as a damp wasp nest, set η =1.

5.

8. The method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The method for performing the spraying operation includes: continuously monitoring the state of the foreign object through a binocular vision module during the spraying process; and immediately analyzing whether there is still a foreign object marker box in the current frame image after a single spraying. If a foreign object marker box is present and the current cumulative number of sprays is less than the preset maximum number of cycles threshold, then execute the next round of spraying; If the foreign object marker disappears or the cumulative number of sprays reaches the threshold, the spraying operation will terminate and the device will be suspended.

9. A method for controlling the removal of foreign objects during power line inspection based on a vehicle-mounted unmanned aerial vehicle (UAV) according to claim 1, characterized in that: The method for energy replenishment includes: after the drone lands, the charging mechanism automatically extends and makes physical contact with the drone's charging contacts, negotiates the charging power with the vehicle power supply through the battery management system (BMS), and executes a constant current followed by a constant voltage charging mode. The system detects the pressure inside the drone's fuel tank. If the pressure is lower than the set value, the solenoid valve is automatically opened, and the fuel is injected into the drone's fuel tank using the pressure inside the vehicle's fuel tank.

10. A power line inspection foreign object removal control system based on a vehicle-mounted drone, used to implement the method as described in any one of claims 1 to 9, characterized in that, include: The vehicle-mounted mobile platform includes an integrated management and control platform, an intelligent hangar, and a first communication module; the integrated management and control platform includes a task planning module, a human-machine interaction module, and a data link module; the intelligent hangar includes a helipad, an automatic charging device, and a fuel replenishment device. The operational drone is a multi-rotor structure equipped with a three-axis self-stabilizing gimbal, a binocular vision imaging module, a foreign object identification and processing module, a distance calculation unit, a multi-degree-of-freedom jet device, a flight control system, and a second communication module. The multi-degree-of-freedom spraying device is mounted on the three-axis self-stabilizing gimbal and includes a spray gun, a high-pressure generator, and a fuel cylinder. The spraying direction of the spray gun is independently controlled by the gimbal. The vehicle-mounted mobile platform and the operating drone establish a data link through the first and second communication modules to realize video stream transmission, control command issuance and status data feedback, and jointly complete the entire process of operation from automatic resupply, autonomous inspection, human-machine collaborative decision-making to adaptive clearing.

Images