An autonomous inspection control system for a vehicle-mounted compound wing unmanned aerial vehicle

Abstract

The present application relates to the technical field of unmanned aerial vehicle inspection, and particularly relates to a vehicle-mounted composite wing unmanned aerial vehicle autonomous inspection control system, comprising a data initialization collection module, a double-target dynamic modeling module and a stop point and task point matching module; the present application collects multiple types of data and unifies coordinates to form initialization input data, defines parameters based on the initialization input data, calculates various time parameters required for modeling, constructs a double-target optimization model with the goal of minimizing invalid flight time and overall inspection total time, and performs multiple rounds of screening on vehicle candidate stop points according to the constraint conditions in the double-target optimization model, assigns task points to optimal stop points to form matching pairs, and realizes the improvement of vehicle-machine cooperation efficiency and operation reliability, and adaptation to complex scene full-autonomous inspection requirements.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) inspection technology, specifically to an autonomous inspection control system for a vehicle-mounted compound wing UAV. Background Technology

[0002] With the rapid development of the low-altitude economy, vehicle-mounted hybrid-wing UAVs, with their dual advantages of vertical take-off and landing and high-speed cruising, have been widely used in long-distance, complex terrain inspection operations in fields such as power, oil and gas, and transportation, gradually replacing traditional manual inspection and fixed UAV inspection modes. Currently, although vehicle-mounted hybrid-wing UAV inspection systems have achieved initial automation, many technical bottlenecks still exist in actual operation, making it difficult to meet the inspection needs of complex scenarios in terms of inspection efficiency, operational safety, and adaptability. Specific problems are as follows: First, there is a lack of scientific and reasonable optimization modeling methods. The existing system has not built an optimization model that fits the vehicle inspection scenario, and cannot quantify the relationship between the candidate vehicle parking points and the drone inspection task points. It also does not clarify the core optimization objectives and constraints, resulting in excessively long invalid flight time of drones and difficulty in controlling the total inspection time. It cannot balance the utilization rate of drone operations with the overall inspection efficiency of the system, and it is difficult to adapt to the high-efficiency operation requirements of long-distance linear inspection scenarios. Secondly, the selection of parking spots and the allocation of task points lack systematicness and precision. The existing system only considers a single condition when selecting candidate parking spots for vehicles, without combining multiple dimensions such as parking feasibility, environmental adaptability, and operational efficiency for multiple rounds of screening. This easily retains invalid parking spots, increasing the computational complexity of subsequent matching. At the same time, the allocation of task points and parking spots lacks a scientific matching mechanism, which can easily lead to problems such as delays in emergency tasks, insufficient drone battery life, duplicate matching, or task omissions. This results in a chaotic inspection process, low operational efficiency, and an inability to achieve efficient vehicle-machine collaborative operations. To address the aforementioned shortcomings, a technical solution is provided. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the existing technology, the present invention provides an autonomous inspection control system for vehicle-mounted compound wing UAVs, which can effectively solve the problems of low inspection efficiency and serious invalid flight of UAVs caused by the lack of optimized modeling and unreasonable matching of docking points and task points in the existing technology.

[0004] To achieve the above objectives, the present invention can be implemented through the following technical solutions: This invention provides an autonomous inspection and control system for a vehicle-mounted compound-wing unmanned aerial vehicle (UAV), comprising: The data initialization and acquisition module is used to collect basic data of the inspection task, vehicle-related data, drone-related data and environmental constraint data, and then convert all data into a unified coordinate system to form the initialization input data. The dual-objective dynamic modeling module, based on the initial input data, sets the inspection task point set and the vehicle candidate parking point set and defines the parameters, calculates various time parameters required for modeling, and then constructs a dual-objective optimization function with the goal of minimizing the invalid flight time and the total overall inspection time, and configures five constraints to form a dual-objective optimization model. The stop-and-task-point matching module is used to perform multiple rounds of screening of candidate vehicle stop points based on the constraints in the dual-objective optimization model, including basic screening, environmental screening, and efficiency screening, to eliminate invalid stop points and form an optimal candidate stop point set. Then, all task points in the inspection task point set are assigned one by one to the corresponding stop points in the optimal candidate stop point set, forming a stop-and-task-point matching pair set.

[0005] Furthermore, the process of defining parameters includes: Each inspection task point in the inspection task point set contains four core parameters: longitude coordinates, latitude coordinates, inspection priority, and required inspection time. Each candidate parking point in the vehicle candidate parking point set contains four core parameters: longitude coordinates, latitude coordinates, parking feasibility indicator, and avionics take-off and landing adaptability.

[0006] Furthermore, various time parameters include: the flight time of the drone from the docking point to the corresponding mission point, the vehicle's travel time, and the invalid flight time; The process of solving various time parameters is as follows: Based on the latitude and longitude coordinates of each candidate vehicle docking point and each inspection task point, calculate the actual flight distance of the UAV from the docking point to the inspection task point, and then combine the preset cruising speed of the UAV to calculate the flight time of the UAV from the docking point to the corresponding task point. Based on the vehicle's current location coordinates and the latitude and longitude coordinates of each candidate stop, the actual distance the vehicle travels from its current location to the candidate stop is calculated. Then, combined with the vehicle's preset speed, the travel time is calculated. The cumulative time of all non-missionary flights after the drone takes off from each candidate docking point is counted as invalid flight time.

[0007] Furthermore, the process of constructing the bi-objective optimization function includes: Using invalid flight time as the first optimization objective, a minimization objective function is constructed. Using the total inspection time as the second optimization objective, a minimization objective function is constructed. The total inspection time is obtained by adding the total flight time of the UAV to and from all docking points and corresponding task points, the total travel time of the vehicle to and from all candidate docking points, and the cumulative inspection time of all inspection task points.

[0008] Furthermore, the five constraints include: priority constraints, drone endurance constraints, vehicle parking constraints, airspace and environment constraints, and mission coverage constraints.

[0009] Furthermore, the basic screening process includes: Remove parking feasibility markers based on vehicle parking constraints. and the adaptability of the avionics for takeoff and landing Candidate stopping points.

[0010] Furthermore, the environmental screening process includes: Candidate docking points located in no-fly zones, dangerous terrain, or where weather conditions exceed the operating range of drones are eliminated based on airspace and environmental constraints.

[0011] Furthermore, the efficiency screening process includes: Calculate the coverage efficiency of each remaining candidate stop after basic screening and environmental screening. The coverage efficiency is calculated as follows: Coverage efficiency = Number of inspection task points that can be covered by the stop + Travel time of the vehicle from the current location to the stop. Based on the calculation results, retain the top 30% of candidate stops in terms of coverage efficiency to form the optimal candidate stop set.

[0012] Furthermore, the process of forming a set of docking point-task point matching pairs includes: Based on the priority constraints, all task points in the inspection task point set are sorted in descending order of inspection priority. For task points of the same priority, they are sorted in descending order of the required inspection time. For each sorted inspection task point, calculate the drone flight distance between the task point and each docking point in the optimal candidate docking point set. Based on the calculation results, select the 3 candidate docking points with the smallest flight distance as potential matching objects for the task point. For each of the three potential matching docking points selected, we will verify whether they meet the drone's endurance constraints. That is, the total flight time of the drone from takeoff to the docking point, after completing the inspection work at the task point and returning to the docking point, shall not exceed 80% of the drone's current remaining endurance. If multiple docking points simultaneously meet the range constraint, the docking point with the highest nacelle takeoff and landing adaptability shall be selected as the final matching docking point for the mission. If none of the three potential matching docking points meet the range constraint, the second-best candidate docking points are re-selected, and the verification process is repeated until a matching docking point that meets the range constraint is found for the task point. Remove the task points that have been matched and assigned from the inspection task point set, and at the same time retain the matching docking point of the task point in the optimal candidate docking point set. Repeat the above process until all inspection task points have been matched and assigned, and finally form a docking point-task point matching pair set.

[0013] The technical solution provided by this invention has the following advantages compared with the known prior art: 1. This invention, by clearly defining the core parameters of inspection task points and vehicle candidate parking points, accurately calculating various time parameters, constructing a dual-objective optimization model with the goals of "minimizing invalid flight time and minimizing the overall inspection time", and configuring five constraints that fit actual operations, thereby significantly shortening the invalid flight time of UAVs, reasonably controlling the overall inspection time, and taking into account both the utilization rate of UAV operations and the overall inspection efficiency of the system, can efficiently adapt to long-distance, complex terrain linear inspection scenarios.

[0014] 2. This invention, through multiple rounds of progressive screening (basic screening, environmental screening, and efficiency screening), combined with constraints such as vehicle parking and airspace environment, eliminates invalid parking points and forms an optimal candidate parking point set, reducing the computational complexity of subsequent matching. At the same time, it sorts according to task priority and combines UAV endurance constraints and UAV nest take-off and landing adaptability, thereby achieving accurate matching between task points and parking points, effectively avoiding delays in emergency tasks, insufficient UAV endurance, duplicate matching, or task omissions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is an overall module block diagram of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1As shown, an autonomous inspection and control system for a vehicle-mounted compound wing UAV includes: a data initialization and acquisition module, a dual-target dynamic modeling module, and a docking point and task point matching module. The data initialization and acquisition module is used to collect basic data of the inspection task, vehicle-related data, drone-related data and environmental constraint data, and then convert all data into a unified coordinate system (such as WGS84 coordinate system) to ensure data consistency and form initial input data. It should be noted that the basic data for the inspection task is distributed by the cloud scheduling platform, which may include the inspection area (latitude and longitude coordinates, boundary thresholds), the set of inspection task points (such as power poles and oil and gas pipeline inspection points, each task point includes a unique ID, latitude and longitude coordinates, inspection priority, inspection duration requirements, and load usage requirements), the total task time limit, and the inspection accuracy requirements. Among them, the inspection priority is divided into three levels (Level 1: Emergency defect review, Level 2: Routine inspection, Level 3: Auxiliary investigation), and the priority directly determines the task execution order. Vehicle-related data is collected in real time by the on-board control station, which may include the vehicle's current location (GPS+RTK fusion positioning, accuracy ±1cm), the range of the vehicle's drivable path (combined with map data, excluding prohibited sections, construction sections, and sections with complex terrain that cannot be passed), the vehicle's speed (0-120km / h, which can be dynamically adjusted according to road conditions), the status of the on-board unit (whether it can take off and land normally, and the status of charging and battery swapping preparation), and the vehicle's remaining driving range. The drone-related data is transmitted back in real time by the drone flight control system, which may include the drone's current status (battery level, attitude, flight speed), maximum endurance, maximum flight radius, minimum space required for take-off and landing, payload operating parameters (such as shooting angle and acquisition frequency), and invalid flight judgment threshold (default: empty flight distance exceeds 1km or empty flight time exceeds 5 minutes, which can be customized according to the scenario). Environmental constraint data is collected by the vehicle-mounted environmental monitoring terminal and may include real-time meteorological data (wind speed, rainfall, visibility), airspace compliance data (no-fly zones, height restrictions) and terrain data (altitude, slope, used to determine the feasibility of vehicle parking and the difficulty of drone flight).

[0019] The dual-objective dynamic modeling module is used to build a dual-objective optimization model based on initial input data. The specific construction process is as follows: First, the core basic data set required for modeling is defined, and the inspection task point set and the vehicle candidate stop point set are set separately. The constituent elements and parameters of each set are defined as follows: The inspection task point set is set as... ,in, These are the labels for the inspection task points. The total number of inspection task points, and each inspection task point in the set. Each contains four core parameters, namely longitude coordinates. Latitude coordinates Inspection priority (Divided into three levels, with smaller values ​​indicating higher priority) and required inspection time. ; Set the set of candidate vehicle stops as , The designation is the number of the candidate parking spot for the vehicle. The total number of candidate parking spots for vehicles, and each candidate parking spot in the set. Each contains four core parameters, namely longitude coordinates. Latitude coordinates docking feasibility sign (Using a binary value rule, a value of 1 indicates that the docking point can be used normally, and a value of 0 indicates that the docking point cannot be used) and the adaptability of the aircraft nest for takeoff and landing. ( A higher value indicates better compatibility between the docking point and the onboard aircraft's takeoff and landing. After setting the two core sets mentioned above, the correlation between candidate vehicle docking points and UAV inspection task points is further quantified. Simultaneously, various time parameters required for modeling are calculated, including the UAV's flight time from the docking point to the corresponding task point, the vehicle's travel time, and invalid flight time. This lays the foundation for constructing the subsequent optimization function. The specific calculation process is as follows: Based on the latitude and longitude coordinates of each candidate vehicle docking point and each inspection task point, the actual flight distance of the UAV from the docking point to the inspection task point is calculated. Then, combined with the UAV's preset cruising speed, the flight time of the UAV from the docking point to the corresponding task point is calculated. ; Based on the vehicle's current location coordinates and the latitude and longitude coordinates of each candidate stop, the actual distance the vehicle travels from its current location to that candidate stop is calculated. Combined with the vehicle's preset speed, the travel time is then calculated. ; The cumulative time of all non-missionary flights after the drone takes off from each candidate docking point is counted, including the idle flight time to and from the docking point and the idle flight time while waiting for operation, etc., and is regarded as invalid flight time. ; After completing the correlation quantification and time parameter calculation, a dual-objective optimization function is constructed to clarify the optimization priority and core objectives, ensuring that the optimization direction aligns with the actual inspection operation requirements: The dual-objective optimization functions are set in descending order of priority, and their specific calculation process is as follows: using ineffective flight time... As the primary optimization objective, construct a minimum objective function. The objective function aims to minimize the drone's flight mileage, ensuring that the drone's flight time is primarily used for inspection tasks, thereby improving drone operational efficiency. Secondly, it uses the total inspection time as the secondary optimization objective, constructing a function to minimize this time. The total time spent on the overall inspection The calculation formula is: In the formula, This refers to the total flight time of the drone between all docking points and the corresponding mission points. This represents the total travel time for the vehicle to travel to and from all candidate stop points. The cumulative inspection time for all inspection task points; When combining, prioritize ensuring the primary optimization objective. The goal is achieved, and then, under the premise of satisfying the first objective and the preset constraints, the second objective is optimized. The two work together to form a complete dual-objective optimization function; This objective function improves the overall efficiency of the inspection system, ensuring that all inspection tasks are completed within the preset total time limit. To ensure the practical feasibility of the constructed dual-objective optimization model and its adaptability to various complex inspection scenarios, the following five constraints are configured simultaneously to reasonably limit the optimization process: a. Priority Constraints: Clearly define the priority matching rights of primary inspection task points, requiring all primary inspection task points to prioritize the matching of docking points and flight paths, and prioritize the allocation of optimal vehicle candidate docking points and UAV flight paths to ensure that emergency inspection tasks can be executed quickly and completed efficiently; b. Drone endurance constraints: To ensure the safety of drone operations, the total flight time of a drone from a candidate docking point to the docking point after completing the inspection work at the corresponding inspection task point must not exceed 80% of the drone's current remaining endurance. 20% of the endurance time is reserved as an emergency backup to deal with sudden flight anomalies. c. Vehicle parking constraints: Strict restrictions are placed on the feasibility of candidate parking spots, requiring each candidate parking spot to simultaneously meet two conditions, namely, parking feasibility indicators. and In other words, it must have docking capability and a nest take-off and landing adaptability of no less than 0.8. At the same time, the candidate docking point must avoid no-fly zones and dangerous terrain to ensure vehicle parking safety and drone take-off and landing safety. d. Airspace and environmental constraints: The flight environment and path of UAVs shall be standardized, requiring that the flight path of UAVs must avoid no-fly zones and altitude-restricted zones. At the same time, the flight environment must meet the standard operating requirements of UAVs, namely, the wind speed during flight shall not exceed 6m / s and the visibility shall not be less than 500m, so as to ensure the stability and safety of UAVs during flight. e. Task Coverage Constraint: Ensure the integrity of inspection tasks, requiring all inspection task points to be matched with the corresponding candidate vehicle parking points, with no task points omitted; at the same time, each inspection task point can only be matched with one optimal candidate parking point to avoid operational chaos and efficiency reduction caused by repeated matching. Through the above process, the construction of the dual-objective optimization model was completed. It not only quantified the correlation between candidate vehicle stopping points and drone inspection task points, but also clarified the optimization objectives and various constraints, providing mathematical support for subsequent path optimization.

[0020] The stop-and-task-point matching module is used to perform multiple rounds of screening of candidate vehicle stop points based on the constraints in the dual-objective optimization model. These screenings include basic screening, environmental screening, and efficiency screening, eliminating invalid stop points and forming the optimal candidate stop point set. This lays the foundation for subsequent dynamic allocation. The specific screening logic is executed in the following order: Based on vehicle parking constraints, a basic screening is conducted to eliminate parking feasibility indicators. (i.e., non-dockable) and nest takeoff and landing adaptability Candidate docking points (i.e., those with insufficient take-off and landing adaptability) are selected to ensure that the selected docking points have basic docking and drone take-off and landing conditions. Then, environmental screening is carried out based on airspace and environmental constraints to further eliminate candidate landing sites located in no-fly zones, dangerous terrain (such as terrain with a slope of >30° that is not suitable for vehicle parking and drone take-off and landing), and weather conditions that exceed the drone's operating range, so as to ensure the safety and feasibility of subsequent operations. Finally, efficiency screening is performed on the remaining candidate stops after the basic and environmental screening. The coverage efficiency of each remaining candidate stop after the first two rounds of screening is calculated as follows: Coverage efficiency = Number of inspection task points that the stop can cover + Travel time of the vehicle from the current location to the stop. Based on the calculation results, the top 30% of candidate stops with the highest coverage efficiency are retained to form the optimal candidate stop set. ,in, , This represents the total number of optimal candidate stops.

[0021] Then, all the task points in the inspection task point set are assigned one by one to the corresponding stops in the optimal candidate stop point set, forming a stop point-task point matching pair set. The specific matching process is as follows: First, the tasks are sorted: based on priority constraints, all task points in the inspection task point set are sorted in order of inspection priority from high to low (level 1 → level 2 → level 3), giving priority to the allocation of level 1 emergency inspection tasks. Among them, for task points of the same priority, they are sorted in order of the inspection time required for each task point from long to short, ensuring that the longer time-consuming tasks are completed and matched first, and avoiding delays in the overall inspection progress due to long time-consuming tasks. Secondly, perform allocation iteration: For each sorted inspection task point, calculate the drone flight distance between the task point and each docking point in the optimal candidate docking point set, and select the 3 candidate docking points with the smallest flight distance as potential matching objects for the task point based on the calculation results. Then, a compatibility check is performed: for each of the three potential matching docking points selected, it is checked whether they meet the drone's endurance constraints, that is, the total flight time of the drone from taking off from the docking point, completing the inspection work at the task point, and returning to the docking point must not exceed 80% of the drone's current remaining endurance. If multiple docking points simultaneously meet the range constraint, the docking point with the highest nacelle takeoff and landing adaptability shall be selected as the final matching docking point for the mission. If none of the three potential matching docking points meet the range constraint, the second-best candidate docking points are re-selected, and the verification process is repeated until a matching docking point that meets the range constraint is found for the task point. Finally, a set update is performed: Task points that have been matched and assigned are removed from the inspection task point set, while their matching docking points are retained in the optimal candidate docking point set. This process of task sorting, assignment iteration, and suitability verification is repeated until all inspection task points have been matched and assigned, ultimately forming a set of docking point-task point matching pairs. Each optimal candidate docking point corresponds to multiple inspection task points, and the specific number is reasonably allocated according to the drone's endurance and the amount of tasks.

[0022] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle-mounted compound-wing unmanned aerial vehicle (UAV) autonomous inspection and control system, comprising a data initialization and acquisition module for collecting basic inspection task data, vehicle-related data, UAV-related data, and environmental constraint data, and then converting all data into a unified coordinate system to form initialization input data; characterized in that, Also includes: The dual-objective dynamic modeling module, based on the initial input data, sets the inspection task point set and the vehicle candidate parking point set and defines the parameters, calculates various time parameters required for modeling, and then constructs a dual-objective optimization function with the goal of minimizing the invalid flight time and the total overall inspection time, and configures five constraints to form a dual-objective optimization model. The stop-and-task-point matching module is used to perform multiple rounds of screening of candidate vehicle stop points based on the constraints in the dual-objective optimization model, including basic screening, environmental screening, and efficiency screening, to eliminate invalid stop points and form an optimal candidate stop point set. Then, all task points in the inspection task point set are assigned one by one to the corresponding stop points in the optimal candidate stop point set, forming a stop-and-task-point matching pair set.

2. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The process of defining parameters includes: Each inspection task point in the inspection task point set contains four core parameters: longitude coordinates, latitude coordinates, inspection priority, and required inspection time. Each candidate parking point in the vehicle candidate parking point set contains four core parameters: longitude coordinates, latitude coordinates, parking feasibility indicator, and avionics take-off and landing adaptability.

3. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, Various time parameters include: the flight time of the drone from the docking point to the corresponding mission point, the vehicle's travel time, and the invalid flight time; The process of solving various time parameters is as follows: Based on the latitude and longitude coordinates of each candidate vehicle docking point and each inspection task point, calculate the actual flight distance of the UAV from the docking point to the inspection task point, and then combine the preset cruising speed of the UAV to calculate the flight time of the UAV from the docking point to the corresponding task point. Based on the vehicle's current location coordinates and the latitude and longitude coordinates of each candidate stop, the actual distance the vehicle travels from its current location to the candidate stop is calculated. Then, combined with the vehicle's preset speed, the travel time is calculated. The cumulative time of all non-missionary flights after the drone takes off from each candidate docking point is counted as invalid flight time.

4. The autonomous inspection control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The process of constructing a bi-objective optimization function includes: Using invalid flight time as the first optimization objective, a minimization objective function is constructed. Using the total inspection time as the second optimization objective, a minimization objective function is constructed. The total inspection time is obtained by adding the total flight time of the UAV to and from all docking points and corresponding task points, the total travel time of the vehicle to and from all candidate docking points, and the cumulative inspection time of all inspection task points.

5. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The five constraints include: priority constraints, drone endurance constraints, vehicle parking constraints, airspace and environment constraints, and mission coverage constraints.

6. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The basic screening process includes: Remove parking feasibility markers based on vehicle parking constraints. and the adaptability of the avionics for takeoff and landing Candidate stopping points.

7. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The environmental screening process includes: Candidate docking points located in no-fly zones, dangerous terrain, or where weather conditions exceed the operating range of drones are eliminated based on airspace and environmental constraints.

8. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The efficiency screening process includes: Calculate the coverage efficiency of each remaining candidate stop after basic screening and environmental screening. The coverage efficiency is calculated as follows: Coverage efficiency = Number of inspection task points that can be covered by the stop + Travel time of the vehicle from the current location to the stop. Based on the calculation results, retain the top 30% of candidate stops in terms of coverage efficiency to form the optimal candidate stop set.

9. The autonomous inspection and control system for a vehicle-mounted composite-wing UAV according to claim 1, characterized in that, The process of forming a set of dock point-task point matching pairs includes: Based on the priority constraints, all task points in the inspection task point set are sorted in descending order of inspection priority. For task points of the same priority, they are sorted in descending order of the required inspection time. For each sorted inspection task point, calculate the drone flight distance between the task point and each docking point in the optimal candidate docking point set. Based on the calculation results, select the 3 candidate docking points with the smallest flight distance as potential matching objects for the task point. For each of the three potential matching docking points selected, we will verify whether they meet the drone's endurance constraints. That is, the total flight time of the drone from takeoff to the docking point, after completing the inspection work at the task point and returning to the docking point, shall not exceed 80% of the drone's current remaining endurance. If multiple docking points simultaneously meet the range constraint, the docking point with the highest nacelle takeoff and landing adaptability shall be selected as the final matching docking point for the mission. If none of the three potential matching docking points meet the range constraint, the second-best candidate docking points are re-selected, and the verification process is repeated until a matching docking point that meets the range constraint is found for the task point. Remove the task points that have been matched and assigned from the inspection task point set, and at the same time retain the matching docking point of the task point in the optimal candidate docking point set. Repeat the above process until all inspection task points have been matched and assigned, and finally form a docking point-task point matching pair set.

Images