An unmanned aerial vehicle cooperative grounding line intelligent monitoring and early warning system

Abstract

This application provides a drone-coordinated intelligent monitoring and early warning system for grounding wires, comprising: real-time acquisition of local environmental data through sensors mounted on multiple drones; extraction of signal interference intensity values ​​and preliminary obstacle location coordinates from the data; generation of a local interference intensity distribution map and an obstacle estimation map for each drone; collaborative fusion of data among multiple drones using a distributed consensus algorithm based on the local interference intensity distribution map and obstacle estimation map to generate a global interference intensity map and a shared obstacle coordinate set; broadcasting shared information from the updated collective safe path set to all drones; enhancing fusion of radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area; acquiring real-time status monitoring data of the grounding wire based on the blind-spot-free mapping; fusing the monitoring data into the global interference intensity map to generate the final monitoring and early warning output.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs). Background Technology

[0002] In the inspection of high-voltage transmission lines and substations, the use of drone swarms plays an indispensable role in maintaining power grid safety. Especially in monitoring and early warning of grounding wire conditions, drones must conduct close-range, real-time inspections of energized high-voltage conductors and crossarms of transmission towers. These work areas are typically very confined and riddled with various high-risk hazards. For example, a loose or broken grounding wire can instantly trigger a short circuit, directly jeopardizing the stable operation of the power system and threatening the safety of nearby residents. Technological innovation in this field not only involves equipment maintenance itself but also profoundly impacts the reliability of public services and the actual effectiveness of emergency response.

[0003] Current inspection methods are clearly inadequate in complex environments. Traditional manual inspections or single-drone operations struggle to effectively address the dynamic changes that constantly occur in high-voltage power line areas, especially small obstacles such as randomly distributed temporary guy wires, remnants of bird nests, or insulator strings swaying in the wind. These objects are often only centimeters in diameter but are hidden within densely packed wires. The detection accuracy and stability of sensing devices are already limited in narrow passages; the added interference from strong electromagnetic fields exacerbates the problem. This interference stems from the combined effects of the magnetic and electric fields generated by high-voltage currents, and its corrosive impact on electronic components is often underestimated. Interference not only significantly reduces the signal-to-noise ratio of lidar or visual sensors but also causes sensor data jitter, blurring the outlines of obstacles.

[0004] From a technical perspective, strong electromagnetic interference has become a core challenge affecting the safety of drone swarm inspections. This interference directly affects the sensing system, causing signal attenuation and distortion. For example, when a drone approaches a 500 kV high-voltage line, the electromagnetic pulse interferes with the millimeter-wave radar's interpretation of the echo signal, causing the error in judging the location of the temporary guy wire ahead to increase from centimeter-level to meter-level, resulting in a deviation in the flight trajectory and thus entering dangerous airspace. At the same time, the interference waves can also infiltrate the attitude control system, disrupting the readings of gyroscopes and accelerometers, causing the drone to roll or pitch instability under conditions of strong winds or insufficient light at night, making it difficult to maintain a hovering attitude. More seriously, electromagnetic noise can block wireless communication links between drones. For example, the signal loss rate of wireless fidelity or long-range radio modules may soar to over 80%, hindering the real-time sharing of hazard information. Once the lead drone encounters interference, the hazard warnings it detects cannot be transmitted to other members of the swarm in a timely manner, and subsequent drones will continue to fly blindly, thus triggering the risk of a chain collision.

[0005] Taking various real-world scenarios as examples, when inspecting the crossarms of transmission towers in rainy or foggy weather, the damp wires can increase the electromagnetic field strength, causing interference that leads to overexposure of visual camera images. Temporary guy wires may be mistakenly identified as background noise, and if a drone recklessly crosses such a section, it could cause a scraping accident. Furthermore, when a swarm passes through a section with multiple transmission towers, if the communication between the lead drone and the trailing drones is interrupted, the entire swarm may veer off course and collide with the grounding wires swaying in the wind, significantly increasing the probability of mission failure.

[0006] These issues reveal a prominent operational contradiction: drone swarms require close collaboration to cover a wide inspection range, but electromagnetic interference leads to isolated sensing and communication disruptions, making it difficult to balance safety and efficiency. Maintaining sensing accuracy and ensuring effective information sharing in such a high-risk environment has become a key bottleneck restricting the successful implementation of inspection missions. Summary of the Invention

[0007] This invention provides an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs), mainly comprising: Local environmental data is collected in real time by sensors on multiple drones. Signal interference intensity values ​​and preliminary obstacle location coordinates are extracted from the data to generate a local interference intensity distribution map and obstacle estimation map for each drone. Based on the local interference intensity distribution map and obstacle estimation map, a distributed consensus algorithm is used to collaboratively fuse data among multiple UAVs to generate a global interference intensity map and a shared obstacle coordinate set. The boundary of the high-interference area is extracted from the global interference intensity map and the shared obstacle coordinate set. The presence of a centimeter-level temporary guy wire within the boundary is detected by millimeter-wave radar. If it exists, the attitude sensor of the UAV is corrected by the Kalman filter algorithm to obtain the corrected attitude data and calculate the stable communication link parameters. The real-time position and velocity vector of each UAV in the cluster are obtained. Based on the corrected attitude data and stable communication link parameters, the dynamic flight envelope boundary is calculated. If the boundary overlaps with the real-time state, the velocity vector is adjusted and path planning is performed to generate an updated collective safe path set. Broadcast shared information from the updated collective safety path set to all drones, and enhance the fusion of radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area; Real-time status monitoring data of the grounding wire is obtained based on the blind-zone-free mapping, and the monitoring data is integrated into the global interference intensity map to generate the final monitoring and early warning output.

[0008] This invention provides an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs), mainly comprising: The local data acquisition and processing module is used to acquire local environmental data in real time through sensors carried by multiple UAVs, extract signal interference intensity values ​​and preliminary obstacle location coordinates from the data, and generate a local interference intensity distribution map and obstacle estimation map for each UAV. The distributed collaborative fusion module is used to collaboratively fuse data from multiple UAVs based on the local interference intensity distribution map and obstacle estimation map using a distributed consensus algorithm, thereby generating a global interference intensity map and a shared obstacle coordinate set. The high interference area detection and correction module is used to extract the boundary of the high interference area from the global interference intensity map and the shared obstacle coordinate set, use millimeter-wave radar to detect whether there are centimeter-level temporary guy wires within the boundary, and if so, use the Kalman filter algorithm to correct the UAV attitude sensor, obtain the corrected attitude data and calculate the stable communication link parameters. The dynamic flight envelope calculation and path planning module is used to obtain the real-time position and velocity vector of each UAV in the cluster, calculate the dynamic flight envelope boundary based on the corrected attitude data and stable communication link parameters, and adjust the velocity vector and perform path planning to generate an updated collective safe path set if the boundary overlaps with the real-time state. The shared information broadcasting and enhanced fusion module is used to broadcast shared information from the updated collective safety path set to all UAVs, and to enhance and fuse radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area. The monitoring data fusion and early warning output module is used to obtain real-time status monitoring data of the grounding wire based on the blind-spot-free mapping, fuse the monitoring data into the global interference intensity map, and generate the final monitoring and early warning output.

[0009] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses an intelligent monitoring and early warning system for grounding wires based on UAV collaborative operations. This system aims to solve a series of interconnected practical problems faced by multi-UAV swarms in complex nighttime inspection scenarios. Such scenarios typically involve high-intensity signal interference and dense obstacles, specifically manifested in inaccurate assessment of interference intensity, insufficient sharing of obstacle coordinate information, communication link instability caused by flight attitude data deviations, multiple blind spots in dynamic path planning, and a lack of real-time monitoring and early warning mechanisms for grounding wire status.

[0010] The system collects local environmental information in real time through deployed sensors, extracting signal interference intensity and obstacle coordinates for the current area, and constructs a local environmental map based on this data. Then, a distributed consensus algorithm is used to collaboratively fuse the local maps acquired by each UAV node, ultimately forming a globally unified interference intensity distribution map and a shared set of obstacle coordinates. Based on this global map, the system can accurately identify and delineate the boundary contours of high-intensity interference areas. Simultaneously, millimeter-wave radar is used to specifically detect slender obstacles such as temporary guy wires that may exist in the environment.

[0011] For flight attitude data, the system employs a Kalman filter algorithm for correction and calculates the parameters required to maintain a stable communication link based on the corrected data. Combining the real-time position and velocity information reported by each UAV, the system dynamically adjusts its flight envelope boundaries, thereby generating an updated set of flight paths. Under nighttime operating conditions, the system also enhances and fuses radar-acquired point cloud data with broadcast information, achieving blind-spot-free panoramic mapping of the inspection area. Finally, the real-time monitoring data of the grounding wire is integrated into the global map, and the system comprehensively processes all information to generate a complete monitoring and early warning output.

[0012] This method significantly improves the accuracy and overall safety of drone swarm inspection tasks, ensuring full coverage of the inspection range and real-time early warning information, thereby effectively optimizing the collective collaborative efficiency of multi-drone systems under complex environmental conditions. Attached Figure Description

[0013] Figure 1 This is a flowchart of an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to the present invention.

[0014] Figure 2This is a schematic diagram of an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to the present invention.

[0015] Figure 3 This is another schematic diagram of an intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to the present invention.

[0016] Figure 4 This is a schematic diagram of the structure of a drone-assisted intelligent monitoring and early warning system for grounding wires according to the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figures 1-4 This embodiment of a drone-assisted intelligent monitoring and early warning system for grounding wires may specifically include: Step S101: Local environmental data is collected in real time by sensors carried by multiple drones. Signal interference intensity values ​​and preliminary obstacle location coordinates are extracted from the data to generate a local interference intensity distribution map and obstacle estimation map for each drone.

[0019] Raw environmental data is collected using sensors mounted on multiple drones. The raw data is filtered and calibrated to obtain signal interference intensity values ​​and preliminary obstacle location coordinates, forming an initial dataset. Based on this initial dataset, the signal interference intensity values ​​are divided into low, medium, and high levels using a binning method, generating a local interference intensity distribution map for each drone. High-interference-level regions are identified from these maps. If the signal interference intensity value in this region exceeds a preset threshold, inverse distance weighted interpolation is used to calculate the precise boundary of the high-interference region. After obtaining the precise boundary of the high-interference region, the preliminary obstacle location coordinates are superimposed and compared with the precise boundary. If the coordinate point is located within the high-interference region, it is marked as a low-confidence obstacle point; otherwise, it is marked as a high-confidence obstacle point, generating an obstacle confidence map for each drone. Based on the obstacle confidence map, the DBSCAN algorithm is used to cluster the high-confidence obstacle points, determining multiple obstacle clusters and obtaining the overall obstacle distribution pattern. Using this overall obstacle distribution pattern, the spatial relationship between the drone swarm's preset flight path and the obstacle clusters is calculated. If an overlap between the path and the area is detected, the path offset is recalculated using the A* algorithm to generate an adjusted flight path. After obtaining the adjusted flight path, subsequent data collection waypoints are assigned to each UAV based on its current position and the adjusted flight path, ensuring that all designated areas are covered by waypoints from at least one UAV.

[0020] Step S102: Based on the local interference intensity distribution map and obstacle estimation map, a distributed consensus algorithm is used to collaboratively fuse the data among multiple UAVs to generate a global interference intensity map and a shared obstacle coordinate set.

[0021] By fusing data from multiple UAVs using a local interference intensity distribution map and obstacle estimation map, a preliminary global interference intensity map is obtained. Based on this preliminary global interference intensity map, a consistency index among the data from multiple UAVs is obtained. If the consistency index is below a preset threshold, the data weights are adjusted to obtain an optimized global interference intensity map. If the consistency index is above the preset threshold, the preliminary global interference intensity map is directly used as the optimized global interference intensity map. Using the optimized global interference intensity map, the Paxos algorithm is used to verify the coordinate consistency among the multiple UAVs. If the consistency meets the requirements, the coordinate data is fused to obtain a final shared obstacle coordinate set. The correlation between the optimized global interference intensity map and the coordinate set is obtained using the final shared obstacle coordinate set, and the correlation strength is determined to obtain integrated map data. Based on the integrated map data, potential conflict areas are extracted. If conflict areas exist, high-risk points are marked to obtain an enhanced global interference intensity map. Using the enhanced global interference intensity map, the final shared obstacle coordinate set is fused to determine the overall environment assessment and obtain comprehensive navigation guidance data.

[0022] Step S103: Extract the boundary of the high interference area from the global interference intensity map and the shared obstacle coordinate set, use millimeter-wave radar to detect whether there is a centimeter-level temporary guy wire within the boundary, if so, use the Kalman filter algorithm to correct the UAV attitude sensor, obtain the corrected attitude data and calculate the stable communication link parameters.

[0023] The boundary coordinates of high-interference areas are extracted from the global interference intensity map. Based on these boundary coordinates, the millimeter-wave radar scans the space within the boundary coordinate range to acquire high-resolution point cloud data. Point cloud data is processed using a point cloud density clustering method to identify linearly dense point clusters. If such clusters are identified, they are recorded as temporary guide lines, and the 3D coordinates of the guide line endpoints are output. The guide line endpoint coordinates and the attitude angle data output in real time by the UAV's inertial measurement unit are input into a Kalman filter. The attitude angles are filtered using the guide line coordinates as observation values, and the corrected attitude angle data is output. Using the corrected attitude angle data, combined with the known positions of the UAV and the ground station, the pitch and azimuth angles along the signal propagation path are calculated. Based on these angles, a preset link attenuation table is consulted to obtain the link stability index. The interference intensity values ​​of each pixel within the high-interference area are read from the global interference intensity map. The interference intensity values ​​are compared with a preset threshold, and the coordinates of pixels exceeding the threshold are marked as critical interference points. The system reads the coordinates of all obstacles identified in the millimeter-wave radar point cloud data, fuses them with the coordinates of key interference points, and uses the A-satellite path search algorithm to generate a local obstacle avoidance path point sequence, constrained by avoiding the coordinates of all obstacles and key interference points. Based on the local obstacle avoidance path point sequence, the flight controller calculates attitude control commands and communication frequency switching commands, while simultaneously receiving real-time point cloud data from the millimeter-wave radar as environmental feedback. Iteratively executing path point sequence tracking completes flight control.

[0024] Step S104: Obtain the real-time position and velocity vector of each UAV in the cluster, calculate the dynamic flight envelope boundary based on the corrected attitude data and stable communication link parameters, and if the boundary overlaps with the real-time state, adjust the velocity vector and perform path planning to generate an updated collective safe path set.

[0025] Real-time position and velocity values ​​are acquired from each drone in the cluster. Simultaneously, corrected attitude angle data and communication link status values ​​are obtained from the drones. Based on the real-time position, velocity, attitude angle data, and communication link status values, a preset safe distance model is used to calculate the dynamic flight envelope boundary of each drone. It is determined whether the real-time position value enters the dynamic flight envelope boundary. If so, the velocity vector of the affected drone is adjusted. Based on the adjusted velocity vector, the A-satellite path planning algorithm is used to generate an updated set of flight paths for the affected drones. The updated flight path set is converted into control commands and issued to the drones in the cluster.

[0026] Step S105: Broadcast shared information from the updated collective safety path set to all UAVs, and enhance and fuse radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area.

[0027] From a pre-defined collective safe flight path database, waypoints and no-fly zone coordinates that all UAVs must know are extracted as shared flight path information. This shared flight path information, along with its real-time updates, is distributed to each UAV in the swarm via a broadcast communication protocol, thus obtaining an initial inspection route for each UAV. Based on the initial inspection route, the 3D coordinate position of the UAV swarm under nighttime conditions is obtained. Point cloud data collected by the airborne millimeter-wave radar is acquired simultaneously. The 3D coordinate position of the UAV swarm is overlaid with the radar point cloud data, and the point cloud is compressed using a voxel grid downsampling tool to generate a preliminary point cloud map of the nighttime environment. Based on this map, the current coverage area of ​​the UAV swarm's sensors is determined. For this coverage area, the newly acquired radar point cloud data is integrated with the terrain data from the shared flight path information. Using an iterative nearest-point algorithm, point cloud data acquired at different times are registered to the same coordinate system. For the registered fused point cloud map, a statistical outlier removal tool is used to filter out noise points, and the 3D spatial volume not covered by point cloud data is calculated. If the uncovered spatial volume exceeds a preset volume threshold, it is determined to be a potential blind spot. By adjusting the flight path library of the UAV swarm, flight paths are replanned for areas with identified blind spots, and new inspection tasks are assigned, resulting in an updated set of UAV coverage flight paths. Based on this updated set, the UAV swarm is driven to fly, and its updated 3D coordinates and corresponding radar point cloud data are acquired in real time. The real-time positions and point cloud data are compared with the fused point cloud map, and the UAV flight paths are dynamically corrected to ensure they cover the target area as planned, thus generating an initial version of the blind-spot-free point cloud map. Using this initial version, combined with regional building distribution data from the geographic information system, morphological closing operations are used to fill in and smooth the point cloud map, resulting in a detailed final inspection coverage point cloud map. If local missing point cloud areas are still detected in the final inspection coverage point cloud map, supplementary scanning commands are sent to the relevant UAVs via broadcast communication protocol. The point cloud data returned by the UAVs after performing the supplementary scan is acquired and fused again with the final inspection coverage point cloud map. An iterative nearest-point algorithm is used for registration to determine and output a complete blind-spot-free 3D environment mapping result.

[0028] Step S106: Obtain real-time status monitoring data of the grounding wire according to the blind-zone-free mapping, integrate the monitoring data into the global interference intensity map, and generate the final monitoring and early warning output.

[0029] A blind-spot-free mapping technique is used to acquire the real-time state flow of the grounding wire. This real-time state flow includes current and voltage data. Based on the real-time state flow, a sliding window is used to calculate the effective current value of each map grid within a time window, which is then used as the monitoring value for that grid. The historical average interference field strength value for each map grid is obtained from a pre-established global interference field strength database. A weighted average algorithm is used to fuse the monitoring value and the interference field strength value, with the weighting coefficients calibrated according to the field environment, to obtain the fused value for each map grid. If the fused value exceeds a preset threshold, the map grid is determined to be abnormal. Based on the location of the abnormal map grid and the percentage by which its fused value exceeds the threshold, a warning level is determined: exceeding the threshold by 100% is a Level 1 warning, and exceeding it by 50% is a Level 2 warning. A monitoring warning output containing the warning level and the coordinates of the abnormal map grid is generated at the output end.

[0030] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A drone-assisted intelligent monitoring and early warning system for grounding wires, characterized in that, The method includes: Local environmental data is collected in real time by sensors on multiple drones. Signal interference intensity values ​​and preliminary obstacle location coordinates are extracted from the data to generate a local interference intensity distribution map and obstacle estimation map for each drone. Based on the local interference intensity distribution map and obstacle estimation map, a distributed consensus algorithm is used to collaboratively fuse data among multiple UAVs to generate a global interference intensity map and a shared obstacle coordinate set. The boundary of the high-interference area is extracted from the global interference intensity map and the shared obstacle coordinate set. The presence of a centimeter-level temporary guy wire within the boundary is detected by millimeter-wave radar. If it exists, the attitude sensor of the UAV is corrected by the Kalman filter algorithm to obtain the corrected attitude data and calculate the stable communication link parameters. The real-time position and velocity vector of each UAV in the cluster are obtained. Based on the corrected attitude data and stable communication link parameters, the dynamic flight envelope boundary is calculated. If the boundary overlaps with the real-time state, the velocity vector is adjusted and path planning is performed to generate an updated collective safe path set. Broadcast shared information from the updated collective safety path set to all drones, and enhance the fusion of radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area; Real-time status monitoring data of the grounding wire is obtained based on the blind-zone-free mapping, and the monitoring data is integrated into the global interference intensity map to generate the final monitoring and early warning output.

2. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process involves real-time acquisition of local environmental data using sensors mounted on multiple drones, extracting signal interference intensity values ​​and preliminary obstacle location coordinates from the data, and generating a local interference intensity distribution map and obstacle estimation map for each drone. This includes: Raw environmental data is collected using sensors mounted on multiple drones; The raw data is filtered and calibrated to obtain the signal interference intensity value and preliminary obstacle location coordinates, forming an initial data set; Based on the initial dataset, the signal interference intensity values ​​are divided into three levels—low, medium, and high—using a binning method to generate a local interference intensity distribution map for each UAV. Identify high-interference-level areas from the local interference intensity distribution map; If the signal interference intensity value in the area exceeds the preset threshold, the inverse distance weighted interpolation method is used to calculate the data of the area to obtain the precise boundary of the high interference area. After obtaining the precise boundary of the high-interference area, the preliminary obstacle location coordinates are superimposed and compared with the precise boundary of the high-interference area; If the coordinate point is located in a high interference area, then the coordinate point is marked as a low confidence obstacle point; otherwise, it is marked as a high confidence obstacle point, and an obstacle confidence map is generated for each UAV. Based on the obstacle confidence map, the DBSCAN algorithm is used to cluster high-confidence obstacle points to determine the clustering areas of multiple obstacles and obtain the overall pattern of obstacle distribution. By analyzing the overall pattern of obstacle distribution, the spatial relationship between the pre-set flight path of the UAV swarm and the obstacle cluster area is calculated. If the path overlaps with the region, the path offset is recalculated using the A* algorithm to generate an adjusted flight path. After obtaining the adjusted flight path, each UAV is assigned subsequent data collection waypoints based on its current location and the adjusted flight path, ensuring that all designated areas are covered by waypoints from at least one UAV.

3. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step involves using a distributed consensus algorithm to collaboratively fuse data from multiple UAVs based on the local interference intensity distribution map and obstacle estimation map, generating a global interference intensity map and a shared obstacle coordinate set, including: By using local interference intensity distribution maps and obstacle estimation maps, and fusing data from multiple UAVs using the Paxos algorithm, a preliminary draft of the global interference intensity map is obtained. Based on the initial draft of the global interference intensity map, obtain the data consistency index among multiple drones. If the consistency index is lower than the preset threshold, adjust the data weight to obtain the optimized global interference intensity map. If the consistency index is higher than the preset threshold, directly use the initial draft of the global interference intensity map as the optimized global interference intensity map. By optimizing the global interference intensity map, the Paxos algorithm is used to verify the coordinate consistency among multiple UAVs. If the consistency requirement is met, the coordinate data is fused to obtain the final set of shared obstacle coordinates. By finally sharing the obstacle coordinate set, the association relationship between the optimized global interference intensity map and the coordinate set is obtained, the association strength is determined, and integrated map data is obtained. Based on the integrated map data, potential conflict areas are extracted. If a conflict area is determined to exist, high-risk points are marked to obtain an enhanced global interference intensity map. By using an enhanced global interference intensity map and fusing the final shared obstacle coordinate set, an overall environmental assessment is determined, resulting in comprehensive navigation guidance data.

4. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process involves extracting high-interference region boundaries from the global interference intensity map and shared obstacle coordinate set, using millimeter-wave radar to detect the presence of centimeter-level temporary guy wires within the boundaries, and if present, using a Kalman filter algorithm to correct the UAV attitude sensor, obtaining corrected attitude data, and calculating stable communication link parameters, including: Extract the boundary coordinates of high-interference areas from the global interference intensity map; Based on the boundary coordinates, the millimeter-wave radar is controlled to scan the space within the boundary coordinate range to obtain high-resolution point cloud data; Point cloud data is processed using a point cloud density clustering method to identify point clusters that are linearly densely distributed in the point cloud. If such point clusters are identified, they are recorded as temporary lines and the three-dimensional coordinates of the line endpoints are output. The coordinates of the end point of the cable and the attitude angle data output in real time by the UAV inertial measurement unit are input into the Kalman filter. The attitude angle is filtered using the cable coordinates as the observation value, and the corrected attitude angle data is output. Using the corrected attitude angle data, combined with the known positions of the UAV and the ground station, the pitch and azimuth angles on the signal propagation path are calculated. Based on these angles, a preset link attenuation table is consulted to obtain the link stability index. Read the interference intensity value of each pixel in the high interference area from the global interference intensity map, compare the interference intensity value with the preset threshold, and mark the coordinates of the pixels that exceed the threshold as key interference points. Read the coordinates of all obstacles identified in the millimeter-wave radar point cloud data, fuse them with the coordinates of key interference points, and use the A-satellite path search algorithm to generate a local obstacle avoidance path point sequence with the constraint of avoiding all obstacle coordinates and key interference point coordinates. Based on the local obstacle avoidance path point sequence, the flight controller calculates attitude control commands and communication frequency switching commands, while receiving real-time point cloud data from millimeter-wave radar as environmental feedback, iteratively executing path point sequence tracking to complete flight control.

5. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of acquiring the real-time position and velocity vectors of each UAV in the cluster, calculating the dynamic flight envelope boundary based on the corrected attitude data and stable communication link parameters, and adjusting the velocity vectors and performing path planning to generate an updated collective safe path set if the boundary overlaps with the real-time state includes: Obtain real-time position and speed values ​​from each drone in the cluster; Simultaneously, the corrected attitude angle data and communication link status values ​​are acquired from the UAV; Based on real-time position, velocity, and attitude angle data, along with communication link status values, a preset safe distance model is used to calculate the dynamic flight envelope boundary of each UAV. Determine whether the real-time position value has entered the dynamic flight envelope boundary; If it enters, adjust the velocity vector of the affected drone; Based on the adjusted velocity vector, the A-satellite path planning algorithm is used to generate an updated set of flight paths for the affected UAVs; The updated set of flight paths is converted into control commands and sent to the drones in the cluster.

6. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step of broadcasting shared information from the updated collective safety path set to all drones, and enhancing and fusing radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area, includes: Extract the waypoints and no-fly zone coordinates that all drones must know from the pre-set collective safe flight path database, and use them as shared flight path information; The shared route information and its real-time updates are distributed to each drone in the drone swarm via a broadcast communication protocol, thereby obtaining the initial inspection route for each drone. Based on the initial inspection route, obtain the three-dimensional coordinate position of the drone swarm under nighttime conditions; Simultaneously acquire point cloud data collected by airborne millimeter-wave radar; The three-dimensional coordinates of the drone swarm are overlaid with radar point cloud data. The point cloud is compressed using a voxel grid downsampling tool to generate a preliminary point cloud map of the nighttime environment. Based on this map, the current coverage area of ​​the drone swarm's sensors is determined. For this coverage area, the newly acquired radar point cloud data will be integrated with the terrain data in the shared flight route information; The iterative nearest point algorithm is used to register point cloud data acquired at different times to the same coordinate system; For the registered and fused point cloud map, use the statistical outlier removal tool to filter out noise points and calculate the 3D spatial volume not covered by point cloud data; If the volume of the uncovered space exceeds a preset volume threshold, it is determined that there is a potential blind spot; By adjusting the flight path library of the drone swarm, the flight paths for the identified blind spots are replanned, and new inspection tasks are assigned, resulting in an updated set of drone coverage flight paths. Based on the updated set of coverage routes, drive the drone swarm to fly and acquire its updated three-dimensional coordinate position and corresponding radar point cloud data in real time; By comparing real-time location data with point cloud data and fused point cloud map, the drone flight path is dynamically corrected to ensure that it covers the target area as planned, thereby generating an initial version of the blind-spot-free point cloud map. Using the initial version of the blind-spot-free point cloud map, combined with regional building distribution data from the geographic information system, the point cloud map is filled and smoothed using morphological closing operation tools to obtain a final inspection coverage point cloud map with complete details. If local missing point cloud areas are still detected in the final inspection coverage point cloud map, a supplementary scanning command is sent to the relevant drones via broadcast communication protocol; The point cloud data returned by the UAV after performing supplementary scanning is acquired and then fused with the final inspection coverage point cloud map. The nearest point algorithm is used for registration to determine and output a complete blind-spot-free 3D environment mapping result.

7. The intelligent monitoring and early warning system for grounding wires in collaboration with unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The step of obtaining real-time status monitoring data of the grounding wire based on the blind-zone-free mapping, integrating the monitoring data into the global interference intensity map, and generating the final monitoring and early warning output includes: The real-time state flow of the grounding wire is obtained using blind-zone-free mapping technology. The real-time state flow includes current and voltage data; Based on the real-time flow of state variables, the effective current value of each map grid within the time window is calculated using a sliding window and used as the monitoring value of that map grid. Obtain the historical average interference field strength value for each map grid from a pre-established global interference field strength database; A weighted average algorithm is used to fuse the monitored values ​​and the interference field intensity values. The weighting coefficients are calibrated and set according to the on-site environment to obtain the fused value for each map grid. If the fusion value exceeds the preset threshold, it is determined that the map grid is abnormal; Based on the location of the abnormal map grid and the percentage of its fusion value exceeding the threshold, the warning level is determined. Exceeding the threshold by 100% is a Level 1 warning, and exceeding the threshold by 50% is a Level 2 warning. The output terminal generates a monitoring and early warning output that includes the warning level and the coordinates of the abnormal map grid.

8. A drone-assisted intelligent monitoring and early warning system for grounding wires, characterized in that, The system includes: The local data acquisition and processing module is used to acquire local environmental data in real time through sensors carried by multiple UAVs, extract signal interference intensity values ​​and preliminary obstacle location coordinates from the data, and generate a local interference intensity distribution map and obstacle estimation map for each UAV. The distributed collaborative fusion module is used to collaboratively fuse data from multiple UAVs based on the local interference intensity distribution map and obstacle estimation map using a distributed consensus algorithm, thereby generating a global interference intensity map and a shared obstacle coordinate set. The high interference area detection and correction module is used to extract the boundary of the high interference area from the global interference intensity map and the shared obstacle coordinate set, use millimeter-wave radar to detect whether there are centimeter-level temporary guy wires within the boundary, and if so, use the Kalman filter algorithm to correct the UAV attitude sensor, obtain the corrected attitude data and calculate the stable communication link parameters. The dynamic flight envelope calculation and path planning module is used to obtain the real-time position and velocity vector of each UAV in the cluster, calculate the dynamic flight envelope boundary based on the corrected attitude data and stable communication link parameters, and adjust the velocity vector and perform path planning to generate an updated collective safe path set if the boundary overlaps with the real-time state. The shared information broadcasting and enhanced fusion module is used to broadcast shared information from the updated collective safety path set to all UAVs, and to enhance and fuse radar point cloud data with the shared information under nighttime conditions to generate a blind-spot-free mapping of the inspection coverage area. The monitoring data fusion and early warning output module is used to obtain real-time status monitoring data of the grounding wire based on the blind-spot-free mapping, fuse the monitoring data into the global interference intensity map, and generate the final monitoring and early warning output.

Images