A method and device for unmanned aerial vehicle inspection control

Abstract

This application provides a method and apparatus for controlling unmanned aerial vehicle (UAV) inspections, relating to the field of collaborative UAV inspections. The method includes: responding to the inspection operations of a first UAV and a second UAV on different inspection routes within a reservoir area on the same target inspection area; acquiring real-time environmental data corresponding to the target inspection area through a three-dimensional data twin model of the reservoir area; inputting the real-time environmental data into a reservoir inspection risk judgment model and determining whether the target inspection area is a key inspection area; if it is confirmed that the target inspection area is not a key inspection area, adjusting the inspection routes of the first UAV and the second UAV according to a preset adjustment method; and controlling the first UAV and the second UAV to inspect the reservoir area according to the adjusted inspection routes. This application solves the problem that existing reservoir inspection methods cannot avoid multiple UAVs repeatedly inspecting the same section, thus increasing unnecessary flight distance and extending the inspection cycle.

Description

Technical Field

[0001] This application relates to the field of drone collaborative inspection, and in particular to a drone inspection control method and device. Background Technology

[0002] In reservoir areas, due to the vast area, undulating terrain, complex distribution of water surfaces, mountains, dams, and power transmission facilities, as well as the large number of inspection targets and their wide spatial span, coupled with the rapid changes in local wind disturbances, sunlight reflection, and communication obstruction over time, a single drone cannot cover all inspection targets within a limited time. Therefore, it is necessary to improve inspection efficiency and ensure the completeness of inspection results by using multiple drones for collaborative inspection.

[0003] Existing technologies mostly employ fixed inspection tracks or multi-UAV scheduling methods based on static task division, achieving collaborative inspection by having multiple UAVs execute preset inspection routes. However, this method cannot dynamically adjust the inspection routes according to real-time changes in the reservoir environment. When there are large water surface reflection areas, areas with concentrated wind disturbances, or areas with densely distributed inspection objects in the reservoir area, multiple UAVs often generate highly overlapping inspection paths in some areas, causing UAVs to repeatedly inspect the same sections, increasing unnecessary flight distances and prolonging the inspection cycle.

[0004] Therefore, there is an urgent need for a method and device for drone inspection and control. Summary of the Invention

[0005] This application provides a method and apparatus for controlling unmanned aerial vehicle (UAV) inspections, which solves the problem that existing warehouse inspections cannot avoid multiple UAVs repeatedly inspecting the same section, thus increasing unnecessary flight distance and extending the inspection cycle.

[0006] The first aspect of this application provides a method for controlling unmanned aerial vehicle (UAV) inspections. The method includes: responding to the inspection operations of a first UAV and a second UAV on different inspection routes in a reservoir area on the same target inspection area; obtaining real-time environmental data corresponding to the target inspection area through a three-dimensional data twin model corresponding to the reservoir area; inputting the real-time environmental data into a reservoir inspection risk discrimination model and determining whether the target inspection area is a key inspection area; if it is confirmed that the target inspection area is not a key inspection area, adjusting the inspection routes of the first UAV and the second UAV according to a preset adjustment method; and controlling the first UAV and the second UAV to inspect the reservoir area according to the adjusted inspection routes.

[0007] Optionally, before responding to the inspection operations of the first and second UAVs on different inspection routes in the reservoir area on the same target inspection area, the method further includes constructing an inspection route corresponding to each UAV. Specifically, this includes: acquiring multi-source environmental data of the reservoir area, and constructing an inspection task corresponding to each UAV based on the multi-source environmental data of the reservoir area; the multi-source environmental data of the reservoir area includes three-dimensional terrain data, reservoir area hydrological environment data, reservoir area fixed risk target data, and reservoir area meteorological disturbance data; constructing an inspection trajectory constraint set according to the task construction elements and UAV operation capability elements associated with the inspection task; the task construction elements include the spatial location of the reservoir area inspection object, the task coverage area, and the task priority; the UAV operation capability elements include the UAV endurance, UAV flight performance, and UAV payload type; and constructing an inspection route corresponding to each UAV based on the inspection trajectory constraint set.

[0008] Optionally, before obtaining real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model corresponding to the reservoir area, the method further includes constructing a three-dimensional data twin model, specifically including: dividing the reservoir area into regions according to the spatial layering rules of the reservoir area, and configuring corresponding multi-source environmental perception sensors for each region; obtaining regional perception data corresponding to each region based on the multi-source environmental perception sensors; and constructing a three-dimensional data twin model through the regional perception data.

[0009] Optionally, real-time environmental data is input into the reservoir inspection risk discrimination model to determine whether the target inspection area is a key inspection area. Specifically, this includes: determining multiple regional risk characteristics based on real-time environmental data, including risk characteristics formed by local wind field disturbances in the reservoir, imaging visibility risk characteristics formed by changes in reservoir illumination and visibility, and navigation and communication risk characteristics formed by RTK signal stability and electromagnetic interference intensity; performing correlation and coupling processing on multiple regional risk characteristics, and calculating a comprehensive risk threshold based on the temporal change relationship, spatial overlap relationship, and nonlinear gain relationship between the risk characteristics of each region; determining whether the comprehensive risk threshold is greater than a preset risk threshold; if it is confirmed that the comprehensive risk threshold is greater than the preset risk threshold, then outputting that the target inspection area is a key inspection area.

[0010] Optionally, after confirming that the comprehensive risk threshold is greater than the preset risk threshold and outputting the target inspection area as the key inspection area, the method further includes: controlling the first UAV and the second UAV to inspect the target inspection area according to the original inspection route while maintaining a cooperative safe distance.

[0011] Optionally, the first and second UAVs are controlled to inspect the reservoir area according to the adjusted inspection routes. Specifically, this includes: acquiring the first inspection route and first operational capability element data corresponding to the first UAV, and the second inspection route and second operational capability element data corresponding to the second UAV; calculating a first adaptation matching value based on real-time environmental data and the first operational capability element data using a UAV inspection adaptation discrimination model, and calculating a second adaptation matching value based on real-time environmental data and the second operational capability element data; determining whether the first adaptation matching value is greater than the second adaptation matching value; if it is confirmed that the first adaptation matching value is greater than the second adaptation matching value, then updating the second inspection route corresponding to the second UAV to the third inspection route, while the first inspection route remains unchanged.

[0012] Optionally, updating the second inspection route corresponding to the second UAV to the third inspection route specifically includes: obtaining the spatial coverage relationship between the first and second inspection routes in the spatial coordinate system of the reservoir area; performing dynamic association processing on the local inspection paths of the second inspection route based on the spatial coverage relationship; determining the target update path in the local inspection path that meets the reservoir inspection path risk constraints based on the dynamic association processing, the reservoir inspection path risk constraints including environmental risk constraints, lateral obstacle avoidance risk constraints, and altitude layer feasibility risk constraints; and adjusting the second inspection route to the third inspection route based on the target update path.

[0013] A second aspect of this application provides a drone inspection control device, which includes an acquisition module and a processing module, wherein... The acquisition module is used to respond to the inspection operations of the first and second UAVs on different inspection routes in the reservoir area on the same target inspection area; and to acquire the real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model of the reservoir area.

[0014] The processing module is used to input real-time environmental data into the reservoir inspection risk judgment model and determine whether the target inspection area is a key inspection area. If it is confirmed that the target inspection area is not a key inspection area, the inspection routes of the first UAV and the second UAV are adjusted according to the preset adjustment method. The first UAV and the second UAV are controlled to inspect the reservoir area according to the adjusted inspection routes.

[0015] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform any of the methods described above.

[0016] A fourth aspect of this application provides a computer-readable storage medium storing a computer program, which is executed by a processor using the method described in any of the foregoing descriptions.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Responding to the inspection operations of the first and second UAVs on different inspection routes in the reservoir area on the same target inspection area; obtaining real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model of the reservoir area; inputting the real-time environmental data into the reservoir inspection risk judgment model and determining whether the target inspection area is a key inspection area; if it is confirmed that the target inspection area is not a key inspection area, adjusting the inspection routes of the first and second UAVs respectively according to the preset adjustment method; controlling the first and second UAVs to inspect the reservoir area according to the adjusted inspection routes, thereby realizing the dynamic coordination of multi-UAV inspection routes in the complex three-dimensional environment of the reservoir area, so that the two inspection routes maintain the integrity and safety of the inspection task while avoiding duplicate inspections, and improve the efficiency of inspection resource utilization and the overall coordination of inspection operations.

[0018] 2. Divide the reservoir area into regions according to the spatial stratification rules of the reservoir area, and configure corresponding multi-source environmental perception sensors for each region; acquire regional perception data for each region based on the multi-source environmental perception sensors; construct a three-dimensional data twin model through the regional perception data, so that the static terrain structure and dynamic environmental state of the reservoir area can be mapped in real time under a unified spatial coordinate system, and the three-dimensional data twin model can simultaneously express the spatial morphological characteristics and environmental change characteristics of each region of the reservoir area, and provide continuous, accurate and updatable basic environmental information for subsequent risk assessment, inspection route adjustment and multi-UAV collaborative control.

[0019] 3. Obtain the spatial coverage relationship between the first and second inspection routes in the spatial coordinate system of the reservoir area; perform dynamic correlation processing on the local inspection paths of the second inspection route based on the spatial coverage relationship; determine the target update path in the local inspection path that meets the reservoir inspection path risk constraints based on the dynamic correlation processing, including environmental risk constraints, lateral obstacle avoidance risk constraints, and altitude-level feasibility risk constraints; adjust the second inspection route to the third inspection route based on the target update path, thereby eliminating path overlap, spatial conflict, and environmental infeasibility of the second inspection route near the target inspection area while maintaining the integrity of the overall inspection task, so that the updated third inspection route can avoid high-risk or low-necessity repetitive sections, and achieve safe, stable, and efficient inspection trajectory planning under the conditions of meeting environmental risk constraints, lateral obstacle avoidance risk constraints, and altitude-level feasibility risk constraints, thereby improving the path separation and operational coordination in the multi-UAV collaborative inspection process. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a drone inspection control method provided in an embodiment of this application; Figure 2 This is a schematic diagram of a module of an unmanned aerial vehicle (UAV) inspection control device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0021] Explanation of reference numerals in the attached figures: 21. Acquisition module; 22. Processing module; 301. Processor; 302. Communication bus; 303. User interface; 304. Network interface; 305. Memory. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0023] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0024] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0025] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] Please refer to Figure 1 The diagram shows a flowchart of a drone inspection control method provided in an embodiment of this application. The flowchart mainly includes the following steps: S101 to S105.

[0027] Step S101, in response to the inspection operations of the first UAV and the second UAV on different inspection routes in the reservoir area on the same target inspection area.

[0028] Specifically, in the inspection scenario of a reservoir area, characterized by its large scale, significant elevation differences, numerous obstacles, and frequent disturbances, multiple drones need to perform high-frequency inspections of objects such as the dam, mountains, spillway, water surface, intake, and shoreline, following different inspection routes. During this process, any two drones may enter the same target inspection area at some point during their respective inspection tasks. The reservoir area's highly complex spatial structure, narrow flight corridors, significant terrain obstruction, and highly time-varying wind disturbances make it prone to collaborative interference, overlapping inspection areas, and even airspace conflicts between different drones in scenarios involving overlapping inspection routes, overlapping inspection areas, and close proximity of inspection paths. Since the inspection routes of different drones are generated independently, and environmental disturbances are random, failure to proactively respond to the inspection behavior of any two drones about to enter the same area may lead to reduced task efficiency, duplicate data collection in local areas, difficulty in compensating for blind spots, and difficulty in timely avoidance of dynamic risks.

[0029] To avoid inspection conflicts, improve inspection continuity, and maintain the integrity of reservoir area inspection tasks during the aforementioned inspection process, it is necessary to trigger step S101 when any two drones (i.e., the first drone and the second drone) are detected to be about to enter the same target inspection area along different inspection routes during the drone inspection process. Potential inspection overlap issues should be identified in advance before the drones arrive. Then, based on the real-time environmental changes of the area to be entered, the status of inspection blind spots, and the differences in the operational capabilities of each drone, it should be determined whether subsequent data acquisition, risk assessment, and inspection route adjustment are necessary, thereby forming a preventive coordination mechanism for inspection conflicts.

[0030] In one possible implementation, step S101 further includes: acquiring multi-source environmental data of the reservoir area, and constructing an inspection task corresponding to each UAV based on the multi-source environmental data of the reservoir area; the multi-source environmental data of the reservoir area includes three-dimensional terrain data, reservoir area hydrological environment data, reservoir area fixed risk target data, and reservoir area meteorological disturbance data; constructing an inspection trajectory constraint set according to the task construction elements and UAV operation capability elements associated with the inspection task; the task construction elements include the spatial location of the reservoir area inspection object, the task coverage area, and the task priority; the UAV operation capability elements include the UAV endurance, UAV flight performance, and UAV payload type; and constructing an inspection route corresponding to each UAV based on the inspection trajectory constraint set.

[0031] Specifically, multi-source environmental data of the reservoir area is acquired, and inspection tasks corresponding to each UAV are constructed based on this data. When constructing the inspection task for each UAV, the acquired multi-source environmental data of the reservoir area is first mapped to a unified spatial coordinate system of the reservoir area. Then, based on the positional accuracy requirements, environmental adaptability requirements, and operational requirements of the inspection task, the roles of multiple UAVs in the task are differentiated. For UAVs responsible for imaging close to the dam surface, the inspection task typically includes a high-precision close-range flight path, a specific zoom window, and a high-coverage-density shooting strategy. For UAVs responsible for large-scale water surface inspection or landslide early warning, the inspection task includes high-speed inspection commands, a large field-of-view shooting plan, and a cross-regional multi-target coverage strategy. By classifying, locating, and prioritizing the inspection objects in the three-dimensional space of the reservoir area, the process of constructing the inspection task can assign a set of tasks suitable for each UAV's capabilities and inspection targets.

[0032] The reservoir area's multi-source environmental data includes 3D topographic data, hydrological environmental data, fixed risk target data, and meteorological disturbance data. 3D topographic data reflects the spatial geometry of the dam, mountains, canyons, water surface reflection zones, water diversion channels, and power transmission towers. By meshing the 3D topography, the passable space, hazardous areas, and minimum safe flight altitude for drones can be clearly defined. Hydrological environmental data describes water surface fluctuations, water level variations, spillway flow intensity, and the interference of water surface reflection on imaging. This data provides safe zone restrictions for close-up water surface photography or waterway inspections by drones. Fixed risk target data includes key areas with dam cracks, landslide-prone areas, high-risk crossings of power transmission lines, and areas with hazardous equipment that should be avoided. This data is used to mark key inspection areas and form a long-term risk basis for path avoidance. The reservoir area meteorological disturbance data includes local wind field disturbances, changes in light intensity, fog distribution range, humidity change trends, and thunderstorm or severe convective weather warnings. This data can provide real-time risk assessment basis for inspection routes, enabling drone inspection routes to be dynamically adjusted before environmental changes occur.

[0033] A set of inspection trajectory constraints is constructed based on the task construction elements and UAV operational capability elements associated with the inspection task. Task construction elements refer to the characteristic data of the reservoir inspection objects that must be associated when constructing the inspection task, including the spatial location of the inspection objects in the reservoir area, the task coverage area, and the task priority. The spatial location of the inspection objects in the reservoir area is used to pinpoint the spatial coordinates of dam crack points, mountain monitoring points, key areas on the water surface, and critical equipment, thereby clarifying the target locations that the inspection path must pass through or cover. The task coverage area represents the minimum coverage volume required for the inspection object in three-dimensional space. For example, dam-fitting inspection requires generating a continuous surface-fitting flight trajectory along the dam surface, while landslide monitoring requires generating a wide-area coverage path across the slope. The task priority is used to determine the order of task execution when multiple inspection objects exist simultaneously. For example, spillway inspection takes precedence over general road inspection, and high-risk mountain areas take precedence over routine vegetation inspection.

[0034] The operational capability elements of unmanned aerial vehicles (UAVs) refer to the performance characteristics used to characterize UAVs when performing inspection tasks, including UAV endurance, flight performance, and payload type. UAV endurance determines the maximum length and duration of the inspection route; for example, UAVs with shorter endurance are suitable for high-precision tasks in localized areas, while those with longer endurance are suitable for cross-regional inspections. UAV flight performance includes flight speed, wind resistance, rate of climb, and braking distance, which determines whether the UAV is suitable for inspection tasks in high-wind areas or narrow corridors. UAV payload type distinguishes between different payload types such as visible light cameras, infrared cameras, LiDAR, or multispectral sensors, ensuring that the inspection route can be adapted to the equipment's imaging angle, focal length range, and target recognition requirements.

[0035] The inspection trajectory constraint set is a set of restrictions formed by the combined effects of mission construction elements and UAV operational capability elements. It is used to constrain the UAV to generate executable, safe, and mission-compliant inspection routes within the reservoir area. The inspection trajectory constraint set includes three aspects: spatial accessibility constraints, mission coverage constraints, and performance feasibility constraints. Spatial accessibility constraints ensure that the UAV's flight path does not enter areas obstructed by mountains or prohibited areas; mission coverage constraints ensure that the inspection route can cover the inspection object and meet minimum imaging requirements; performance feasibility constraints restrict the inspection route to be generated within the limits allowed by the UAV's endurance, wind resistance, and payload field of view. For example, when the UAV payload is a high-zoom visible light camera, the inspection route should be as close as possible to the dam or mountain surface; when the UAV's wind resistance is weak, it should avoid planning routes into canyon areas with strong wind disturbances.

[0036] Inspection routes are constructed for each UAV based on the inspection trajectory constraint set. When constructing the inspection routes, the constraints included in the inspection trajectory constraint set are applied jointly to the 3D path planning algorithm to generate flight paths for each UAV in the 3D space of the reservoir area that meet coverage, safety, and performance requirements. Taking dam inspection as an example, when constructing a flight path that fits the dam, the path needs to unfold along the curved surface of the dam, while avoiding the dam shoulder area with strong wind disturbances, and ensuring the UAV's endurance while covering the entire length of the dam. Taking landslide monitoring as an example, a lateral scanning route needs to be generated on the mountain slope, while maintaining sufficient lateral obstacle avoidance margin and avoiding a decrease in flight stability caused by excessively drastic changes in altitude.

[0037] Step S102: Obtain real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model of the reservoir area.

[0038] Specifically, real-time environmental data corresponding to the target inspection area is obtained through a three-dimensional data twin model of the reservoir area, and environmental information reflecting the current real state of the reservoir area is provided to the UAV before it performs the inspection task, so as to support the subsequent determination of key inspection areas and adjustment of inspection routes.

[0039] In one possible implementation, step S102 further includes: dividing the reservoir area into regions according to the spatial stratification rules of the reservoir area, and configuring a corresponding multi-source environmental sensing sensor for each region; acquiring regional sensing data corresponding to each region based on the multi-source environmental sensing sensor; and constructing a three-dimensional data twin model through the regional sensing data.

[0040] Specifically, the reservoir area is divided into regions according to the spatial stratification rules. These rules divide reservoir areas with different topographic features, hydrological attributes, and meteorological disturbances into several spatial sub-regions with relatively consistent environmental attributes, facilitating the subsequent establishment of environmental data mapping relationships in three-dimensional space. These spatial stratification rules typically consider the three-dimensional topographic elevation differences, water surface distribution, mountain slope variations, dam structure, the locational relationship between the intake and spillway, and the layout of fixed risk targets such as power transmission lines. For example, for reservoir areas with significant elevation differences, mountains can be divided into slope top, mid-slope, and slope bottom areas according to topographic contour lines; for water surface areas, upstream, midstream, and downstream water surface sub-regions can be formed based on water surface reflection intensity, flow velocity variations, or water width; and for dam structures, separate zones can be created for the dam face, abutments, and bottom. Through these spatial stratification rules, the entire reservoir area is transformed into manageable, labelable, and sustainably updated spatial sub-regions in three-dimensional space.

[0041] After the area is divided, corresponding multi-source environmental sensing sensors are configured for each area. The sensor configuration process determines the sensor type and deployment location based on the area's spatial altitude, terrain unevenness, wind disturbance intensity, light intensity variation frequency, and communication obstruction. For mountain slopes, due to wind shear and localized strong gusts, wind speed and direction sensors and IMU-assisted sensing devices are required to capture real-time wind disturbances. For water surfaces, due to drastic changes in light reflection, light intensity sensors, visibility sensors, or water surface reflection imaging devices are required to describe the current imaging quality. For dam areas and their vicinity, since structures may obstruct communication signals, RTK base station quality monitoring devices, electromagnetic interference monitoring modules, or GNSS signal quality sensors are required. For valley areas prone to fog accumulation, humidity sensors, temperature sensors, and fog density monitoring equipment can be configured. For example, in a mountain slope area near a dam where the wind field is stable but the light intensity changes drastically, light and visibility sensors should be prioritized; while in a canyon wind gap area where the wind direction fluctuates drastically, wind speed and direction sensors and IMU analysis modules should be configured to obtain more timely wind field disturbance data.

[0042] Regional sensing data for each area is acquired using multi-source environmental sensing sensors. This data characterizes the actual environmental state of the area at the current moment, typically including local wind speed, wind direction, wind field variation trends, light intensity, visibility level, water surface reflection parameters, humidity, temperature, fog density, RTK signal quality, electromagnetic interference intensity, and other environmental disturbance parameters. The regional sensing data consists of raw or pre-processed time-series data directly output from the sensors. Spatially, it corresponds to a sub-grid dividing the area, and temporally, it is updated at a fixed time sequence, providing a continuous and dynamic real-time description of the environmental conditions in different sub-areas of the reservoir area in three-dimensional space.

[0043] A 3D data twin model is constructed using regional perception data. The construction process begins by spatially aligning the regional perception data of each sub-region according to the reservoir's 3D coordinate system, ensuring each data record corresponds to a regional attribute point in 3D space. Then, based on the spatial adjacency relationships between sub-regions, the terrain elevation data of the regions, and the risk category of each sub-region, all regional attribute points are connected to form a continuous 3D environmental field. Finally, the regional perception data is fused with the reservoir's basic terrain data, ensuring the 3D model maintains the realism of the terrain geometry while reflecting the changing local environmental state over time. This 3D data twin model not only includes static 3D spatial structural features but also possesses a dynamically updated environmental state layer. This allows UAVs to accurately acquire environmental information of the target inspection area at the current moment during inspections, providing reliable input data for subsequent risk assessment and path adjustment.

[0044] Step S103: Input real-time environmental data into the reservoir inspection risk discrimination model and determine whether the target inspection area is a key inspection area.

[0045] In one possible implementation, step S103 further includes: determining multiple regional risk characteristics based on real-time environmental data, including risk characteristics formed by local wind field disturbances in the reservoir, imaging visibility risk characteristics formed by changes in reservoir illumination and visibility, and navigation and communication risk characteristics formed by RTK signal stability and electromagnetic interference intensity; performing correlation coupling processing on the multiple regional risk characteristics, and calculating a comprehensive risk threshold based on the temporal change relationship, spatial overlap relationship, and nonlinear gain relationship between the regional risk characteristics; determining whether the comprehensive risk threshold is greater than a preset risk threshold; if it is confirmed that the comprehensive risk threshold is greater than the preset risk threshold, then outputting the target inspection area as a key inspection area.

[0046] Specifically, firstly, basic risk components are calculated for wind field disturbance, imaging visibility, and navigation and communication status within the target inspection area. By normalizing the local wind speed sequence, wind direction changes, and turbulence intensity, the wind field disturbance risk component is obtained, denoted as:

[0047] in, The wind field disturbance risk component is used to characterize the degree of influence of local wind field on the attitude and trajectory stability of UAV; The root mean square value of local wind speed within a preset time window reflects the wind speed intensity in that area. The wind speed is set as a reference for wind field risk based on the safe flight experience or historical accident data of drones in the reservoir area. It is the absolute value of the wind speed gradient along the direction perpendicular to the main flight direction, used to characterize the degree of drastic spatial variation of wind speed; This is a reference value for the wind speed gradient, used to normalize the gradient term; The characteristic time constant of turbulence calculated in this region can be obtained based on the decay rate of the wind speed autocorrelation function and is used to characterize turbulent instability. The reference time constant for turbulence; The wind field risk mapping coefficient is obtained by fitting the correspondence between historical wind fields and drone disturbance events. It is used to balance the contributions of wind speed intensity, wind speed gradient and turbulent instability to the risk components. These are monotonically increasing nonlinear mapping functions, such as sigmoid functions, used to compress linear combination results into a defined interval, making the risk components more sensitive under extreme wind conditions.

[0048] To address image visibility, an imaging perturbation component is constructed using illumination intensity, illumination gradient, visibility distance, and camera exposure compensation parameters, resulting in an image visibility risk component: in, The imaging visibility risk component is used to characterize the difficulty for drones to acquire clear images in the current area; The variance of light intensity within a preset time window is used to reflect the instability of light in time or space, such as the intense flickering caused by water surface reflection. This serves as a reference value for the variance of illumination, used to normalize the degree of illumination fluctuation. This is the equivalent visibility distance estimated based on fog density or visibility sensors, which decreases as fog worsens or smoke concentration increases. This is a reference value for the visible distance; It represents the total absolute change in camera exposure compensation parameters within the same time window, reflecting the frequency with which the imaging system adjusts to maintain image brightness. Adjust the reference value for exposure; The imaging risk mapping coefficient is determined by fitting the image quality evaluation results in the reservoir scene.

[0049] For navigation and communication, risk components are constructed using RTK correction success rate, changes in satellite visibility, and electromagnetic interference power density, resulting in: in, This is a navigation and communication risk component, used to characterize the difficulty for a drone to maintain a stable position and communication link in the area; To improve the success rate of RTK correction within a preset window, The difference between the expected reference correction success rate and the expected success rate is used to characterize the degree of decline in positioning stability. This represents the maximum change in the number of visible satellites within a window, used to reflect rapid fluctuations in satellite visibility. This serves as a reference value for changes in the number of visible satellites. This refers to the electromagnetic interference power density or equivalent interference index measured in this area. This is a reference value for interference power. The navigation communication risk mapping coefficient is determined by analyzing the correspondence between abnormal communication events and the above parameters.

[0050] After obtaining the three risk components, to introduce the coupling effect in the time and space dimensions, the risk components of the current moment and the historical moments are constructed into an extended risk vector, and the time rate of change and spatial gradient terms are added to form a spatiotemporal embedding vector: in, It is a spatiotemporal embedding vector used to centrally represent the risk status of the target inspection area at the current moment and its spatiotemporal evolution characteristics; These are the rates of change of the three risk components over a preset time window. They can be obtained by differentiating the risk components over two or more consecutive frames and are used to reflect whether the risk is worsening or mitigating. These are the regional gradients of the three risk components in space. They can be calculated based on the differences in risk components between the target inspection area and several neighboring areas, and are used to characterize whether the area is on the edge of a high-risk area or in a steep gradient zone.

[0051] Traditional algorithms typically use a simple weighting method to fuse the aforementioned risks. However, in this implementation, the comprehensive risk threshold is generated by applying a nonlinear quadratic discriminant model to the spatiotemporal embedding vector, specifically: in, The comprehensive risk threshold is used to ultimately characterize the overall risk level of the target inspection area at the current moment; The discriminant matrix is ​​symmetrical and is obtained by training samples of "normal area" and "high-risk area" in historical inspection data. Its off-diagonal elements are used to characterize the cross-influence relationship between different risk components and their time change rate and spatial gradient, thereby reflecting the coupling effect of wind field, imaging and navigation communication in a complex environment. This is a linear discrimination vector used to adjust the basic contribution of each risk component to the discrimination result; This is a bias term used to adjust the overall discrimination threshold; It remains a monotonic nonlinear function, used to compress the output interval and enhance sensitivity to high-risk samples. This is achieved by introducing a quadratic term. This means that the overall risk depends not only on the magnitude of a single risk component, but also on the combination relationship between different risk components and their spatiotemporal derivatives, thus significantly different from the linear weighting method.

[0052] Given a preset risk threshold In the case that, when the following conditions are met At that time, the target inspection area is identified as a key inspection area. A preset risk threshold is set. It can be obtained by statistically analyzing samples of "dangerous areas" and "safe areas" marked in historical inspection records, or it can be set according to the safety strategy of the reservoir operation and maintenance unit.

[0053] When a target inspection area is identified as a key inspection area, due to the high risk or mission importance of the area at the current moment, there is no need to adjust the inspection route to avoid duplicate inspections. Instead, the original inspection strategy is prioritized, and multiple drones simultaneously inspect the area to ensure the integrity and stability of the inspection results and achieve full coverage of high-risk areas. In the inspection task system composed of multiple drones, when the multiple drones are designated as first and second drones, the first and second drones are controlled to inspect the target inspection area according to the original inspection route while maintaining a cooperative safety distance. This includes synchronously retrieving the spatial trajectory of the original inspection route in the 3D data twin scene, and determining the feasible cooperative flight interval based on the minimum spatial distance between the two inspection routes within the target inspection area. The horizontal and vertical intervals between the two drones are dynamically adjusted based on the altitude layer planning results corresponding to the original inspection route, flight direction, and real-time attitude parameters of the two drones, ensuring that the two drones are within the target inspection area. The flight trajectories consistently maintain spatial separation above the minimum cooperative safety distance. During inspections, the flight speed, heading angle, and acceleration of both the first and second UAVs are corrected in real time to ensure synchronized operation without altering the original inspection route structure. This guarantees consistent imaging time windows for key inspection areas, resulting in multi-view, multi-angle synchronous inspections. Simultaneously, the cooperative safety distance between the two UAVs is fine-tuned based on real-time environmental data changes. When wind disturbances intensify, lighting changes drastically, or navigation and communication risks increase, the cooperative safety distance automatically increases to enhance inspection stability. Furthermore, real-time feedback on the sensor status, attitude stability, and local wind resistance of both UAVs during inspections allows them to maintain a dual-path enhanced coverage in key areas along the original inspection route. This achieves a resource-saving inspection processing mode that ensures effective inspection of key areas without increasing path adjustment costs.

[0054] Step S104: If it is confirmed that the target inspection area is not a key inspection area, the inspection routes of the first UAV and the second UAV are adjusted according to the preset adjustment method.

[0055] Specifically, given a preset risk threshold In the case that, when the following conditions are met If the target inspection area is confirmed to be a non-key inspection area, it means that there is no risk or importance in this area that requires simultaneous coverage by multiple drones at the current moment. Therefore, there is no need to continue deploying multiple drones to conduct repeated inspections of this area. In this case, to improve overall inspection efficiency and avoid unnecessary resource consumption, the inspection routes of the first and second drones are adjusted according to a preset adjustment method. By replacing, shifting, or reversing the timing of local path segments of the two inspection routes near the target inspection area, the two drones can avoid route overlap or inspection conflicts in this area without reducing the integrity of the mission, thereby making the execution of the inspection mission more efficient and coordinated.

[0056] Step S105: Control the first and second UAVs to inspect the reservoir area according to the adjusted inspection route.

[0057] Specifically, by performing spatial consistency verification, time rhythm matching, and linkage correction with real-time environmental data on the updated inspection route, the two UAVs can complete their respective inspection tasks at different altitudes, in different flight directions, or in different time windows, thereby ensuring overall inspection efficiency and safety.

[0058] In one possible implementation, step S105 further includes: acquiring the first inspection route and first operational capability element data corresponding to the first UAV, and the second inspection route and second operational capability element data corresponding to the second UAV; calculating a first adaptation matching value based on real-time environmental data and the first operational capability element data using a UAV inspection adaptation discrimination model, and calculating a second adaptation matching value based on real-time environmental data and the second operational capability element data; determining whether the first adaptation matching value is greater than the second adaptation matching value; if it is confirmed that the first adaptation matching value is greater than the second adaptation matching value, then updating the second inspection route corresponding to the second UAV. The third inspection route is adopted, while the first inspection route remains unchanged. The second inspection route corresponding to the second UAV is updated to the third inspection route. Specifically, this includes: obtaining the spatial coverage relationship between the first and second inspection routes in the spatial coordinate system of the reservoir area; performing dynamic association processing on the local inspection paths of the second inspection route based on the spatial coverage relationship; determining the target update path in the local inspection path that meets the reservoir inspection path risk constraints based on the dynamic association processing. The reservoir inspection path risk constraints include environmental risk constraints, lateral obstacle avoidance risk constraints, and altitude layer feasibility risk constraints; and adjusting the second inspection route to the third inspection route based on the target update path.

[0059] Specifically, the first inspection route and first operational capability element data corresponding to the first UAV, and the second inspection route and second operational capability element data corresponding to the second UAV are obtained. To determine which of the two UAVs is more suitable for undertaking the inspection task near the target inspection area, a UAV inspection suitability discrimination model is introduced. This model calculates the suitability matching value by combining real-time environmental data with the operational capability elements of each UAV. The operational capability element data describes the core operational performance characteristics of the UAV during the inspection task and is a description of the UAV's own capabilities corresponding to the real-time environmental data. Its content is consistent with the "UAV operational capability elements" used when constructing the inspection trajectory constraint set, and is reused in step S105 as input to the inspection suitability discrimination model to determine the suitability of different UAVs for undertaking the task in the updated inspection task. The operational capability data mainly includes three dimensions: UAV endurance, flight performance, and payload type. Endurance describes the maximum distance and time the UAV can fly continuously under current battery conditions, which is an important condition for whether the UAV can cover the updated inspection route. Flight performance describes parameters such as the UAV's speed stability, maximum wind resistance, maximum climb rate, maximum sink rate, maximum horizontal acceleration, and maneuverability. This set of parameters determines whether the UAV can successfully pass through areas with large altitude changes, strong wind disturbances, or dense lateral obstacles in the complex terrain environment of the reservoir. Payload type describes the payload equipment currently carried by the UAV, such as visible light cameras, infrared cameras, LiDAR, multispectral cameras, or other mission-specific sensors. Payload type not only determines the types of inspection tasks the UAV can perform, but also directly affects the payload adaptability score in the inspection adaptability discrimination model, because the inspection targets in different inspection areas have different perception requirements. For example, high-resolution visible light payloads are preferred for dam crack inspection, while LiDAR is more relied upon for early monitoring of landslides. The adaptability matching value is constructed through the following improved algorithm to achieve dynamic quantification of UAV mission suitability in complex reservoir environments. The fit matching value is defined as:

[0060] in, This is the fit matching value; A monotonically increasing function is used to normalize the results; The environmental adaptability score is a description of the controllability of the UAV under real-time environmental data conditions, including wind resistance and visual imaging stability. Its value is based on the real-time wind field disturbance risk component. With imaging visibility risk component The reverse mapping; The performance adaptability score for drones is composed of endurance, maneuverability, mission radius, and speed stability, and is calculated based on drone performance parameters and path geometry (including climb angle, turning radius, and flight distance). The load adaptability score describes the adaptability between the type of sensors carried by the drone and the requirements for identifying the objects being inspected. For example, drones equipped with LiDAR are more suitable for inspecting mountain slopes. The three weights were obtained by training with historical reservoir inspection data. It is a nonlinear coupling function used to describe the interaction between the three types of adaptability scores in complex environments. Its high value indicates that the UAV still has stable execution capabilities in harsh environments, such as when the wind resistance is strong and the imaging quality is stable, resulting in enhanced coupling.

[0061] Calculations were performed on the first and second UAVs respectively. and If satisfied If this condition is met, it means that the first UAV is more suitable as the execution subject in the target area. In this case, it is only necessary to update the second inspection route corresponding to the second UAV and generate a third inspection route, while keeping the first inspection route unchanged; if the condition is met... If this condition is met, it means the second drone is more suitable as the execution subject in the target area. In this case, the first inspection route should be changed while the second inspection route remains unchanged. If the first and second UAVs are adapted to the target inspection area in terms of real-time environmental conditions, operational capabilities, and load adaptability, then both can complete the inspection task of the area with the same reliability and stability. In this case, either the first inspection route or the second inspection route can be changed at will.

[0062] The specific process of changing the inspection route is as follows: First, the first and second inspection routes are spatially aligned within a unified spatial coordinate system of the reservoir area. Spatial coverage is constructed by matching the node sequences, flight direction vectors, and altitude layer distributions of the two routes. This spatial coverage is used to characterize overlapping sections, closely approximating sections, and potential conflict sections with insufficient minimum spacing between paths outside and adjacent to the target inspection area. Based on the spatial coverage, dynamic association processing is performed on the local inspection paths of the second inspection route. During this process, local path segments in the second inspection route that may cause inspection overlap, spatial conflict, or interference are extracted as candidate path segments. Simultaneously, a reference path is established for dynamic association by combining the forward flight direction, path curvature changes, and altitude layer distribution of the first inspection route, thus constructing a geometric matching relationship between the candidate local path segments and the reference path.

[0063] After completing the dynamic correlation processing, each candidate local path segment is input into the reservoir inspection path risk constraint for feasibility screening. The reservoir inspection path risk constraint consists of environmental risk constraints, lateral obstacle avoidance risk constraints, and height-level feasibility risk constraints. The environmental risk constraint is determined by the wind field disturbance risk component in the real-time environmental data. With imaging visibility risk component Candidate local path segments are evaluated to ensure that the environmental controllability of the path segment in the area is within an acceptable range. Lateral obstacle avoidance risk constraints are determined based on the minimum lateral safe distance between the candidate path segment and surrounding mountains, dam structures, water surface reflection areas, and power transmission facilities to prevent the path segment from entering flight hazard zones with narrow spaces or structural obstructions. Altitude-level feasibility risk constraints are determined by comparing the altitude changes of the candidate path segment with the three-dimensional terrain profile, the UAV's maximum climb rate, and the minimum flight altitude requirements to ensure that the path segment has continuity and feasibility in the altitude dimension. Finally, the candidate path segment that is satisfied by all three risk constraints is determined as the target update path.

[0064] The second inspection route is structurally adjusted based on the target update path. Local path segments that previously caused path overlap or spatial conflicts in the second inspection route are replaced with the target update path. After the replacement, the node sequence, flight direction, and path curvature are smoothed to ensure that there are no abrupt changes, flight direction reversals, or altitude jumps at the adjustment points in the third inspection route. This generates a third inspection route that maintains the coverage, structural continuity, and environmental feasibility of the inspection task, enabling the second UAV to avoid the target inspection area and successfully perform the remaining inspection tasks without affecting the overall inspection efficiency.

[0065] Please refer to Figure 2 This illustration shows a schematic diagram of a drone inspection control device according to an embodiment of this application. The device includes an acquisition module 21 and a processing module 22, wherein... The acquisition module 21 is used to respond to the inspection operations of the first UAV and the second UAV on different inspection routes in the reservoir area on the same target inspection area; and to acquire the real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model corresponding to the reservoir area.

[0066] The processing module 22 is used to input real-time environmental data into the reservoir inspection risk discrimination model and determine whether the target inspection area is a key inspection area; if it is confirmed that the target inspection area is not a key inspection area, the inspection routes of the first UAV and the second UAV are adjusted according to the preset adjustment method; and the first UAV and the second UAV are controlled to inspect the reservoir area according to the adjusted inspection routes.

[0067] In one possible implementation, before the acquisition module 21 performs inspection operations on the same target inspection area on different inspection routes in the reservoir area in response to the first UAV and the second UAV in the reservoir area, the method further includes constructing an inspection route corresponding to each UAV. Specifically, this includes: acquiring multi-source environmental data of the reservoir area, and constructing an inspection task corresponding to each UAV based on the multi-source environmental data of the reservoir area; the multi-source environmental data of the reservoir area includes three-dimensional terrain data, reservoir area hydrological environment data, reservoir area fixed risk target data, and reservoir area meteorological disturbance data; constructing an inspection trajectory constraint set according to the task construction elements and UAV operation capability elements associated with the inspection task; the task construction elements include the spatial location of the reservoir area inspection object, the task coverage area, and the task priority; the UAV operation capability elements include the UAV endurance, UAV flight performance, and UAV payload type; and constructing an inspection route corresponding to each UAV based on the inspection trajectory constraint set.

[0068] In one possible implementation, before acquiring the real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model corresponding to the reservoir area, the acquisition module 21 further includes constructing a three-dimensional data twin model, specifically including: dividing the reservoir area into regions according to the spatial layering rules of the reservoir area, and configuring corresponding multi-source environmental perception sensors for each region; acquiring the region perception data corresponding to each region based on the multi-source environmental perception sensors; and constructing a three-dimensional data twin model through the region perception data.

[0069] In one possible implementation, the processing module 22 is used to input real-time environmental data into the reservoir inspection risk discrimination model and determine whether the target inspection area is a key inspection area. Specifically, this includes: determining multiple regional risk characteristics based on real-time environmental data, including risk characteristics formed by local wind field disturbances in the reservoir, imaging visibility risk characteristics formed by changes in reservoir illumination and visibility, and navigation and communication risk characteristics formed by RTK signal stability and electromagnetic interference intensity; performing correlation coupling processing on the multiple regional risk characteristics, and calculating a comprehensive risk threshold based on the temporal change relationship, spatial overlap relationship, and nonlinear gain relationship between the regional risk characteristics; determining whether the comprehensive risk threshold is greater than a preset risk threshold; and if it is confirmed that the comprehensive risk threshold is greater than the preset risk threshold, then outputting that the target inspection area is a key inspection area.

[0070] In one possible implementation, after the processing module 22 outputs the target inspection area as a key inspection area if it is confirmed that the comprehensive risk threshold is greater than the preset risk threshold, the method further includes: controlling the first UAV and the second UAV to inspect the target inspection area according to the original inspection route while maintaining a cooperative safe distance.

[0071] In one possible implementation, the processing module 22 is used to control the first UAV and the second UAV to inspect the reservoir area according to the adjusted inspection route. Specifically, this includes: acquiring the first inspection route and first operational capability element data corresponding to the first UAV, and the second inspection route and second operational capability element data corresponding to the second UAV; calculating a first adaptation matching value based on real-time environmental data and the first operational capability element data using a UAV inspection adaptation discrimination model, and calculating a second adaptation matching value based on real-time environmental data and the second operational capability element data; determining whether the first adaptation matching value is greater than the second adaptation matching value; if it is confirmed that the first adaptation matching value is greater than the second adaptation matching value, then updating the second inspection route corresponding to the second UAV to the third inspection route, while the first inspection route remains unchanged.

[0072] In one possible implementation, the processing module 22 is used to update the second inspection route corresponding to the second UAV to the third inspection route, specifically including: obtaining the spatial coverage relationship between the first inspection route and the second inspection route in the spatial coordinate system of the reservoir area; performing dynamic association processing on the local inspection paths of the second inspection route based on the spatial coverage relationship; determining the target update path in the local inspection path that meets the reservoir inspection path risk constraints based on the dynamic association processing, the reservoir inspection path risk constraints including environmental risk constraints, lateral obstacle avoidance risk constraints, and altitude layer feasibility risk constraints; and adjusting the second inspection route to the third inspection route based on the target update path.

[0073] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0074] This application also provides an electronic device. (See reference...) Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: at least one processor 301, at least one communication bus 302, a user interface 303, at least one network interface 304, and a memory 305.

[0075] The communication bus 302 is used to enable communication between these components.

[0076] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0077] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0078] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0079] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. (Refer to...) Figure 3 The memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a drone inspection and control application.

[0080] exist Figure 3 In the illustrated electronic device, the user interface 303 is primarily used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call the UAV inspection control application stored in the memory 305. When executed by one or more processors 301, the electronic device performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0081] This application also provides a computer-readable storage medium storing instructions. When executed by one or more processors, these instructions cause an electronic device to perform one or more of the methods described in the above embodiments.

[0082] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0083] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0084] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0086] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0087] The above description is merely an exemplary embodiment disclosed in this application and should not be construed as limiting the scope of this application. Any equivalent changes and modifications made in accordance with the teachings of this application shall still fall within the scope of this application.

[0088] This application is intended to cover any variations, uses, or adaptations disclosed herein that follow the general principles disclosed herein and include common knowledge or customary technical means in the art that are not described in this application.

Claims

1. A method for controlling unmanned aerial vehicle (UAV) inspections, characterized in that, The method includes: In response to the inspection operations of the first and second UAVs on different inspection routes in the reservoir area, targeting the same inspection area; Real-time environmental data corresponding to the target inspection area is obtained through a three-dimensional data twin model of the reservoir area. The real-time environmental data is input into the reservoir inspection risk discrimination model to determine whether the target inspection area is a key inspection area. If it is confirmed that the target inspection area is not the key inspection area, the inspection routes of the first UAV and the second UAV are adjusted according to the preset adjustment method respectively; The first and second drones are controlled to inspect the reservoir area according to the adjusted inspection route.

2. The method according to claim 1, characterized in that, Before responding to the inspection operations of the first and second UAVs on different inspection routes in the reservoir area on the same target inspection area, the method further includes constructing an inspection route corresponding to each UAV, specifically including: Acquire multi-source environmental data of the reservoir area, and construct inspection tasks corresponding to each UAV based on the multi-source environmental data of the reservoir area; the multi-source environmental data of the reservoir area includes three-dimensional terrain data, reservoir area hydrological environment data, reservoir area fixed risk target data, and reservoir area meteorological disturbance data; An inspection trajectory constraint set is constructed based on the task construction elements and UAV operation capability elements associated with the inspection task; the task construction elements include the spatial location of the inspection object in the storage area, the task coverage area, and the task priority; the UAV operation capability elements include the UAV endurance, UAV flight performance, and UAV payload type. The inspection route for each UAV is constructed based on the inspection trajectory constraint set.

3. The method according to claim 1, characterized in that, Before obtaining the real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model corresponding to the reservoir area, the method further includes constructing the three-dimensional data twin model, specifically including: The reservoir area is divided into regions according to the spatial stratification rules of the reservoir area, and a corresponding multi-source environmental sensing sensor is configured for each region after division. The area perception data corresponding to each area is obtained based on the multi-source environmental perception sensor. The three-dimensional data twin model is constructed using the region-aware data.

4. The method according to claim 1, characterized in that, The step of inputting the real-time environmental data into the reservoir inspection risk discrimination model and determining whether the target inspection area is a key inspection area specifically includes: Based on the real-time environmental data, multiple regional risk characteristics are determined. These multiple regional risk characteristics include risk characteristics formed by local wind field disturbances in the reservoir, imaging visibility risk characteristics formed by changes in reservoir illumination and visibility, and navigation and communication risk characteristics formed by RTK signal stability and electromagnetic interference intensity. Multiple regional risk features are correlated and coupled, and a comprehensive risk threshold is calculated based on the temporal variation relationship, spatial overlap relationship and nonlinear gain relationship among the regional risk features. Determine whether the overall risk threshold is greater than a preset risk threshold; If it is confirmed that the comprehensive risk threshold is greater than the preset risk threshold, then the target inspection area is output as the key inspection area.

5. The method according to claim 4, characterized in that, After confirming that the comprehensive risk threshold is greater than the preset risk threshold, and then outputting the target inspection area as the key inspection area, the method further includes: Control the first drone and the second drone to inspect the target inspection area according to the original inspection route while maintaining a safe cooperative distance.

6. The method according to claim 1, characterized in that, The control of the first and second drones to inspect the reservoir area according to the adjusted inspection route specifically includes: Acquire the first inspection route and first operational capability element data corresponding to the first UAV, and the second inspection route and second operational capability element data corresponding to the second UAV; The first adaptation matching value is calculated based on the real-time environmental data and the first operational capability element data using the UAV inspection adaptation discrimination model, and the second adaptation matching value is calculated based on the real-time environmental data and the second operational capability element data. Determine whether the first fit matching value is greater than the second fit matching value; If it is confirmed that the first fit matching value is greater than the second fit matching value, then the second inspection route corresponding to the second UAV is updated to the third inspection route, while the first inspection route remains unchanged.

7. The method according to claim 6, characterized in that, The update of the second inspection route corresponding to the second UAV to the third inspection route specifically includes: Obtain the spatial coverage relationship between the first inspection route and the second inspection route in the spatial coordinate system of the reservoir area; Based on the spatial coverage relationship, the local inspection paths of the second inspection route are dynamically associated. Based on the dynamic association processing, the target update path that meets the reservoir inspection path risk constraints in the local inspection path is determined. The reservoir inspection path risk constraints include environmental risk constraints, lateral obstacle avoidance risk constraints, and height layer feasibility risk constraints. Based on the target update path, the second inspection route is adjusted to the third inspection route.

8. A drone inspection control device, characterized in that, The device includes an acquisition module and a processing module, wherein, The acquisition module is used to respond to the inspection operations of the first UAV and the second UAV on different inspection routes in the reservoir area on the same target inspection area; and to acquire the real-time environmental data corresponding to the target inspection area through the three-dimensional data twin model corresponding to the reservoir area. The processing module is used to input the real-time environmental data into the reservoir inspection risk discrimination model and determine whether the target inspection area is a key inspection area; if it is confirmed that the target inspection area is not the key inspection area, the inspection routes of the first UAV and the second UAV are adjusted according to a preset adjustment method; and the first UAV and the second UAV are controlled to inspect the reservoir area according to the adjusted inspection routes.

9. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1 to 7.

Images