Unmanned aerial vehicle autonomous inspection and task planning method under global intelligent platform

Abstract

The invention relates to the technical field of city management, in particular to an unmanned aerial vehicle autonomous inspection and task planning method under a global intelligent platform. Comprising the steps that an inspection task instruction from a global intelligent platform is received, and the inspection task instruction comprises a target inspection area, an inspection precision requirement and inspection time limitation; dividing the target inspection area into a plurality of sub-areas, and calculating the inspection complexity coefficient of each sub-area according to the geographical environment, terrain height and obstacle distribution condition of each sub-area; based on the inspection complexity coefficient, the inspection precision requirement and the inspection time limit, utilizing a preset task allocation algorithm to allocate the inspection task to a plurality of unmanned aerial vehicles; for the sub-region tasks allocated to each unmanned aerial vehicle, generating an optimal flight path by using a path optimization algorithm in combination with the cruising ability, the flight speed and the load weight of the unmanned aerial vehicle; according to the invention, the problems of unreasonable task allocation and low overall efficiency in the existing multi-unmanned aerial vehicle inspection system can be solved.

Description

Technical Field

[0001] The present invention relates to the field of urban management technology, and more specifically to a method for autonomous inspection and mission planning of unmanned aerial vehicles (UAVs) under a global intelligent interconnection platform. Background Art

[0002] With the rapid development of the low-altitude economy, drones are increasingly being used in urban and rural governance, emergency management, power inspections, and other fields. In complex inspection scenarios, a single drone often struggles to meet the demands of large-scale, high-efficiency operations, making collaborative multi-drone operations an inevitable trend. However, existing multi-drone inspection systems exhibit significant flaws in task allocation, primarily due to the fact that task allocation strategies fail to fully account for differences in environmental complexity within the inspection area and the performance of drones. Traditional technologies typically employ simple equal division of regions or fixed path allocation, ignoring the impact of factors such as the geographical environment, terrain height, and obstacle distribution on inspection difficulty. For example, in inspections in urban-rural fringe areas, which encompass both densely populated urban areas with high-rise buildings and densely wooded suburbs, adopting a unified task allocation strategy results in drones spending excessively long operating times in complex areas and increasing idle time in simpler areas, resulting in low overall inspection efficiency. At the same time, existing technologies do not make differentiated considerations on drone performance parameters such as endurance, flight speed, and payload weight, which can easily cause some high-performance drones to be underloaded, while low-performance drones may be overloaded or even return mid-flight due to battery depletion, seriously affecting the integrity and timeliness of inspection tasks.

[0003] Based on the above problems, there is an urgent need for an intelligent task allocation technology that can comprehensively consider the environmental complexity of the inspection area and the performance differences of drones, so as to solve the problems of unreasonable task allocation and overall low efficiency in the existing multi-drone inspection system, and improve the intelligence level and operational efficiency of drone inspections under the global intelligent connection platform. Summary of the Invention

[0004] The purpose of the present invention is to address the shortcomings of the prior art and propose a method for autonomous inspection and mission planning of drones under a global intelligent connection platform, including: Receive inspection task instructions from the global intelligent connection platform, which include the target inspection area, inspection accuracy requirements, and inspection time limit; Divide the target inspection area into multiple sub-areas, and calculate the inspection complexity coefficient of each sub-area based on the geographical environment, terrain height, and obstacle distribution of each sub-area; Based on the inspection complexity coefficient, inspection accuracy requirement and inspection time limit, the inspection task is assigned to multiple drones using a preset task allocation algorithm; For each sub-regional mission assigned to a drone, the optimal flight path is generated using a path optimization algorithm, taking into account the drone's endurance, flight speed, and payload weight. During the drone's inspection mission, the drone's flight status data and mission execution progress data are collected in real time and fed back to the global intelligent connection platform; When a drone malfunctions or encounters an emergency during a mission, the global intelligent connection platform reallocates inspection tasks and adjusts flight paths based on the status of the remaining drones and the mission execution status.

[0005] Preferably, the step of dividing the target inspection area into multiple sub-areas is specifically as follows: according to the shape and area of ​​the target inspection area, the target inspection area is divided into multiple square grid sub-areas of equal size using a grid division method; for each sub-area, its corresponding satellite map data, terrain height data and historical meteorological data are obtained to construct a sub-area environmental information database.

[0006] Further preferably, the step of calculating the inspection complexity coefficient of each sub-area is specifically as follows: Determine the number of buildings in the sub-area based on the sub-area environmental information database , percentage of tree cover area , average terrain slope ; The inspection complexity coefficient is calculated using the following formula: : ; in, is the area of ​​the subregion, 、 、 is the weight coefficient, and , 、 、 Both are greater than 0.

[0007] Further preferably, the preset task allocation algorithm specifically includes: Get the collection of drones participating in the inspection mission , each drone The battery life is , the maximum flight speed is ; For each sub-region , according to its inspection complexity coefficient ,area And the preset inspection time standard per unit area , calculate the expected inspection time of the sub-area ; With the goal of minimizing the time difference in task completion of all drones, a task allocation model is established: ; in, Indicates the allocation to the drone The task allocation model is solved by genetic algorithm to obtain the optimal task allocation solution.

[0008] Further preferably, the path optimization algorithm is specifically: For each drone Assigned sub-region set , taking the geometric center of the sub-region as the path node; Considering the UAV's turning radius restrictions, no-fly zone restrictions, and flight altitude restrictions, a path search space is constructed; An improved algorithm is used to search for paths and introduce a heuristic function: ; in, is the current path node, is the target path node, is the Euclidean distance from the current node to the target node, 、 is the weight coefficient, and , 、 are greater than 0, For sub-region During the search process, the path is dynamically adjusted based on the real-time meteorological data and traffic control data.

[0009] Further preferably, when the improved algorithm is used for path search, the node expansion cost function is: ; in, From the starting node to the current node The actual cost is calculated by the following formula: in, 、 are the coordinates of two adjacent path nodes, is the average flight speed of the UAV.

[0010] Further preferably, the flight status data of the drone includes position coordinates , flight speed , flight attitude angle , Remaining battery power ; The task execution progress data includes the number of sub-areas that have been inspected , the amount of data that has been collected ; The global intelligent platform calculates the task execution risk index of the UAV according to the flight state data and the task execution progress data through the following formula :

[0011] wherein, is the total power of the UAV battery, is the time that has been flown, is the planned flight time, is the expected amount of collected data, , , is a weight coefficient, and , , , are all greater than 0; when the task execution risk index exceeds a preset threshold, a task re-allocation and path adjustment mechanism is triggered.

[0012] Further preferably, the step of the global intelligent platform re-allocation of the inspection task and adjustment of the flight path is specifically: the unfinished task of the UAV that has failed or encountered an unexpected situation is re-allocated to the remaining normally operating UAVs according to the task allocation algorithm; for each UAV after the task is re-allocated, the optimal flight path is re-generated according to the path optimization algorithm, and the new flight path is sent to the corresponding UAV through the wireless communication module.

[0013] Further preferably, it further includes: before the UAV executes the inspection task, the task allocation algorithm and the path optimization algorithm are trained and optimized by using historical inspection data and simulation data; the task allocation algorithm and the path optimization algorithm in the global intelligent platform are periodically updated in version to adapt to different inspection task requirements and environmental changes.

[0014] Further preferably, the global intelligent platform and the UAV perform data transmission through a 5G communication network or a dedicated wireless communication link, and an end-to-end encryption technology is adopted in the data transmission process to ensure the security and integrity of the data transmission.

[0015] Technical effects: The application divides the target inspection area into sub-regions scientifically and calculates the complexity coefficient, and combines the performance of the unmanned aerial vehicle with the task requirements, uses a preset algorithm to perform task allocation and path planning, and forms a complete closed-loop management system. This technology creatively considers the environmental and equipment factors, and forms a significant difference with the traditional simple allocation method. In view of the problems of unreasonable task allocation and low efficiency in the background art, the scheme realizes balanced allocation of tasks, avoids uneven load of the unmanned aerial vehicle, optimizes path planning, reduces energy consumption and risk, and greatly improves the overall efficiency, reliability and intelligent level of the unmanned aerial vehicle inspection under the global intelligent platform. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The flow chart of the unmanned aerial vehicle autonomous inspection and task planning method under the global intelligent platform of the present application. DETAILED DESCRIPTION

[0018] In the following description, specific details are set forth in order to provide a thorough understanding of the embodiments of the present application. However, it should be apparent to those skilled in the art that the present application can be practiced without these specific details. In other instances, well-known systems, devices, circuits and methods have been described in detail to avoid unnecessary detail to avoid obscuring the description of the present application.

[0019] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, integers, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, operations, elements, components and / or sets thereof.

[0020] Please refer to Figure 1, like the traditional technical solutions have the following technical problems: the traditional unmanned aerial vehicle inspection has the problems of unreasonable task allocation, lack of comprehensive consideration in path planning, and inability to effectively respond to abnormalities. For example, when inspecting a large area, if the tasks are not scientifically divided, some unmanned aerial vehicles may be overloaded, and some may be idle, resulting in low overall inspection efficiency; if the path planning only considers the distance without considering environmental factors, the unmanned aerial vehicle may frequently encounter obstacles, affecting task execution; when encountering failures or unexpected situations, the lack of effective mechanisms may cause task interruption or delay. Based on this, the embodiment provides a method for autonomous inspection and task planning of unmanned aerial vehicles under a global intelligent platform, which includes: receiving an inspection task instruction from the global intelligent platform, the inspection task instruction including a target inspection area, an inspection accuracy requirement, and an inspection time limit; dividing the target inspection area into multiple sub-areas, and calculating the inspection complexity coefficient of each sub-area according to the geographical environment, terrain height, and obstacle distribution of each sub-area; based on the inspection complexity coefficient, the inspection accuracy requirement, and the inspection time limit, using a preset task allocation algorithm to allocate the inspection task to multiple unmanned aerial vehicles; for each sub-area task allocated to each unmanned aerial vehicle, combining the endurance, flight speed, and load weight of the unmanned aerial vehicle to generate an optimal flight path using a path optimization algorithm; in the process of executing the inspection task by the unmanned aerial vehicle, real-time collection of flight state data and task execution progress data of the unmanned aerial vehicle, and feedback of the data to the global intelligent platform; when the unmanned aerial vehicle encounters a failure or an unexpected situation during task execution, the global intelligent platform re-allocates the inspection task and adjusts the flight path according to the state and task execution of the remaining unmanned aerial vehicles.

[0021] It is worth mentioning that: the embodiment constructs a complete process of autonomous inspection and task planning of unmanned aerial vehicles under a global intelligent platform. First, the task instruction including the target inspection area, the accuracy, and the time limit is received, then the target area is divided into sub-areas and the complexity coefficient is calculated, the task is allocated according to the coefficient, the accuracy, and the time limit, the optimal path is generated for the unmanned aerial vehicle, the data is collected and fed back in real time during execution, and the task is re-allocated and the path is adjusted when a failure or an unexpected situation occurs. This process covers all links of task receiving, analysis, allocation, execution monitoring, and exception handling, forming a closed-loop management system.

[0022] The technical effects achieved by the above embodiments include: by dividing the area to calculate the complexity coefficient, the task can be reasonably distributed according to the actual environment, the load of each unmanned aerial vehicle is balanced, and the overall inspection efficiency is improved; the optimal path is generated in combination with multiple factors, the invalid path and obstacle avoidance time in the flight of the unmanned aerial vehicle are reduced, the energy consumption is reduced, and the inspection accuracy is improved; the real-time data acquisition feedback and abnormal processing mechanism ensure that the task can still proceed smoothly under various sudden situations, improve the completion rate and reliability of the inspection task, and provide a stable and efficient solution for the unmanned aerial vehicle inspection under the global intelligent platform.

[0023] The traditional technical solution has the following technical problems: in the unmanned aerial vehicle inspection task, if the target area is not divided scientifically, it will lead to uneven task allocation and difficulty in adapting to the inspection needs in different environments. The previous area division method may be too general and does not fully consider the influence of actual geographical environment and meteorological factors on the inspection. For example, in a complex terrain or a region with variable weather, planning a path without understanding these information may cause the unmanned aerial vehicle to be in danger or unable to obtain effective data. Therefore, the step of dividing the target inspection area into multiple sub-areas is: according to the shape and area of the target inspection area, the target inspection area is divided into multiple square grid sub-areas of equal size by using the grid division method; for each sub-area, obtain its corresponding satellite map data, terrain height data and historical weather data, and construct a sub-area environment information database.

[0024] It is worth mentioning that: the embodiment further refines the step of dividing the target inspection area into sub-areas, divides the target area into square grid sub-areas of equal size by using the grid division method, and ensures the standardization and consistency of the division. At the same time, for each sub-area, satellite map data, terrain height data and historical weather data are collected to construct a sub-area environment information database. The database integrates multiple data sources and provides comprehensive data support for subsequent analysis of the characteristics of the sub-area.

[0025] The technical effects achieved by the above embodiments include: the standardized grid division makes the task allocation more operational and quantifiable, facilitating accurate calculation and management of the task of each sub-area. The establishment of the sub-area environment information database enables the system to fully understand the characteristics of each sub-area, providing accurate data basis for subsequent calculation of inspection complexity coefficient, task allocation and path planning. Based on these data, the task allocation can be more in line with the actual environment, the path planning is more safe and efficient, the accuracy and effectiveness of the inspection are improved, and the risk of the unmanned aerial vehicle executing the task in a complex environment is reduced.

[0026] The traditional technical solutions have the following technical problems: the existing technology lacks quantitative standards and comprehensive considerations when evaluating the inspection area difficulty, resulting in inaccurate matching of task allocation and path planning to actual difficulty. Different environmental factors have different effects on unmanned aerial vehicle inspection. Without quantitative analysis, complex areas and simple areas may be treated equally, leading to unreasonable task allocation. For example, building-dense areas and open areas have large differences in requirements for unmanned aerial vehicle flight. Not considering these differences will affect the inspection efficiency and safety. Based on this, the step of calculating the inspection complexity coefficient of each sub-area is specifically: determining the number of buildings in the sub-area based on the sub-area environment information database , tree coverage area proportion , average terrain slope ; The inspection complexity coefficient is calculated by the following formula : ; Wherein, is the area of the sub-area, , , is a weight coefficient, and , , , are all greater than 0.

[0027] The formula is used to calculate the inspection complexity coefficient of the sub-area , and the core purpose is to quantify the complexity of the sub-area for unmanned aerial vehicle inspection tasks, providing an important basis for subsequent task allocation and path planning. In the formula, represents the number of buildings in the sub-area. The presence of buildings will have a multi-faceted impact on the flight of the unmanned aerial vehicle, such as increasing the difficulty of flight path planning, the unmanned aerial vehicle needs to avoid buildings to prevent collision, and buildings may also affect signal transmission, increasing the instability of communication. By dividing the number of buildings by the area of the sub-area , the proportion of the number of buildings per unit area is obtained, reflecting the density of buildings in the sub-area. Tree coverage area proportion is represented by the tree coverage area proportion, which reflects the influence of vegetation in the sub-area on unmanned aerial vehicle inspection. Trees may block the camera view of the unmanned aerial vehicle, resulting in incomplete inspection data collection; in the flight process, the unmanned aerial vehicle also needs to avoid trees, which undoubtedly increases the complexity and risk of flight operation. This parameter directly presents the influence of tree coverage on the inspection task in the form of proportion, directly showing the influence of tree coverage on the inspection task. To average terrain slope, changes in terrain slope affect the drone's flight attitude control and energy consumption. In areas with steeper slopes, the drone consumes more energy to maintain stable flight, and flight speed and path planning also need to be adjusted accordingly. Average terrain slope comprehensively reflects the complexity of the sub-area's terrain. 、 、 is the weight coefficient and satisfies , they are all greater than 0. These weight coefficients are used to adjust the influence of each factor on the inspection complexity coefficient. In different application scenarios and requirements, the importance of the number of buildings, the proportion of tree cover area and the average terrain slope may be different. For example, when inspecting in urban areas, the impact of the number of buildings may be more prominent, in which case it can be appropriately increased. For inspections in forested areas, the weight of tree cover should be increased. By flexibly adjusting the weight coefficients, the calculated inspection complexity coefficient can be made more accurate in real-world situations, enabling more precise task planning and resource allocation.

[0028] It is worth mentioning that this embodiment clarifies the specific method for calculating the sub-area inspection complexity coefficient. , percentage of tree cover area , average terrain slope And other key environmental factors, through the formula Calculate the inspection complexity coefficient, which introduces the weight coefficient 、 、 , the influence of each factor on complexity can be adjusted according to actual needs.

[0029] The technical effects achieved by the above-described embodiments include: the inspection complexity coefficient calculated using this formula can intuitively reflect the inspection difficulty of each sub-area. Task allocation based on this coefficient can assign high-difficulty areas to higher-performing drones, or combine multiple moderately difficult areas for more scientific and reasonable task allocation. During path planning, safer and more efficient paths can also be selected based on the complexity coefficient, avoiding risky drone flights in complex areas. This quantitative assessment method improves the accuracy and scientific nature of task planning, enhancing the overall effectiveness and safety of drone inspections.

[0030] For example, traditional technical solutions have the following technical problems: Traditional drone task allocation methods are mostly simple and crude, such as equal allocation by region, without considering drone performance differences and task difficulty. This can easily cause some drones to run out of power prematurely or have task backlogs, resulting in low overall inspection efficiency. In addition, the lack of scientific objective functions and solution algorithms makes it difficult to find the optimal allocation solution. Based on this, the preset task allocation algorithm specifically includes: acquiring a set of unmanned aerial vehicles participating in a patrol task , each unmanned aerial vehicle has a flight endurance of and a maximum flight speed of ; For each sub-region , according to its patrol complexity coefficient , area and a preset unit area patrol time standard , the estimated patrol time of the sub-region is calculated ; a task allocation model is established to minimize the task completion time difference of all unmanned aerial vehicles: ; wherein represents a set of sub-regions allocated to unmanned aerial vehicle , and the task allocation model is solved by a genetic algorithm to obtain an optimal task allocation scheme.

[0031] This formula is the objective function of the task allocation model, and its core goal is to minimize the task completion time difference of all unmanned aerial vehicles to achieve balanced allocation and efficient execution of the multi-unmanned aerial vehicle cooperative patrol task. In the formula, represents the number of unmanned aerial vehicles participating in the patrol task, is used to index each unmanned aerial vehicle, and all unmanned aerial vehicles are traversed from 1 to . represents a set of sub-regions allocated to unmanned aerial vehicle , i.e. the combination of sub-regions to be patrolled by the th unmanned aerial vehicle. represents the estimated patrol time of sub-region , which takes into account factors such as the complexity and area of the sub-region. represents the sum of the estimated patrol times of all sub-regions undertaken by unmanned aerial vehicle , which reflects the task load of each unmanned aerial vehicle. represents the maximum value in the sum of the estimated patrol times of all unmanned aerial vehicles, i.e. the estimated working time of the unmanned aerial vehicle with the heaviest task load; represents the minimum value, i.e. the estimated working time of the unmanned aerial vehicle with the lightest task load. By calculating The maximum difference of all unmanned aerial vehicle task completion times is obtained, and the objective of the entire formula is to minimize this difference. This means that during task allocation, the system will try to make the task load of each unmanned aerial vehicle tend to be balanced, avoiding the situation that some unmanned aerial vehicles complete the task too early and idle, and some unmanned aerial vehicles have too heavy tasks, resulting in the overall inspection time being lengthened. For example, in a large-area urban inspection task involving multiple unmanned aerial vehicles, task allocation is performed through the objective function, which can reasonably arrange the inspection range of each unmanned aerial vehicle according to the actual situation of each sub-region and the performance of the unmanned aerial vehicle, so that all unmanned aerial vehicles can complete the task at the same time as much as possible, thereby improving the inspection efficiency, fully utilizing the unmanned aerial vehicle resources, and reducing the overall inspection cost.

[0032] It is worth mentioning that: the embodiment details the preset task allocation algorithm. First, the set of unmanned aerial vehicles participating in the inspection task and their endurance time, maximum flight speed and other parameters are obtained, then the estimated inspection time of each sub-region is calculated, and then a task allocation model is established with the objective of minimizing the difference in task completion time of all unmanned aerial vehicles, and the optimal task allocation scheme is obtained by genetic algorithm. The algorithm comprehensively considers the performance of the unmanned aerial vehicle and the task demand of the sub-region, and realizes the reasonable allocation of the task.

[0033] The technical effects achieved by the above embodiment include: the inspection complexity coefficient calculated by the formula can intuitively reflect the inspection difficulty of each sub-region. Based on this coefficient, the region with high difficulty can be allocated to the unmanned aerial vehicle with better performance, or multiple regions with moderate difficulty can be combined for allocation, so that the task allocation is more scientific and reasonable. When planning the path, a safer and more efficient path can be selected according to the complexity coefficient to avoid the unmanned aerial vehicle flying in a complex region. This quantitative evaluation method improves the accuracy and scientificity of task planning, and improves the overall efficiency and safety of unmanned aerial vehicle inspection.

[0034] The traditional technical solution has the following technical problems: the traditional path planning algorithm usually only considers the shortest distance and ignores factors such as terrain and environmental complexity, resulting in a path that may not be able to be actually executed or is inefficient. In actual inspection, dynamic factors such as weather changes and traffic control will also affect the feasibility of the path, and existing algorithms lack dynamic adjustment capability. The present application aims to solve the problem of lack of comprehensive consideration and dynamic adaptability in path planning. Based on this, the path optimization algorithm is specifically: For each unmanned aerial vehicle The set of sub-regions allocated to it The geometric center of the sub-region is taken as the path node; The turning radius limit of the unmanned aerial vehicle, the no-fly area limit and the flight height limit are considered to construct the path search space; An improved algorithm is used for path search and a heuristic function is introduced: ; wherein, is the current path node, is the target path node, is the Euclidean distance from the current node to the target node, , is the weight coefficient, and , , are all greater than 0, is the inspection complexity coefficient of the sub-area ; in the search process, the path is dynamically adjusted according to the real-time acquired meteorological data and traffic control data.

[0035] This formula is the objective function of the task allocation model, and its core goal is to minimize the time difference of all UAV task completion to achieve balanced allocation and efficient execution of multi-UAV cooperative inspection tasks. In the formula, represents the number of UAVs participating in the inspection task, is used to index each UAV, from 1 to all UAVs are traversed. represents the set of sub-areas allocated to UAV , i.e., the combination of sub-areas responsible for inspection by the th UAV. represents the estimated inspection time of sub-area , which takes into account factors such as the complexity and area of the sub-area. represents the total estimated inspection time of all sub-areas tasks undertaken by UAV , which reflects the task load of each UAV. represents the maximum value in the total estimated inspection time of all UAV tasks, i.e., the estimated working time of the UAV with the heaviest task load; represents the minimum value, i.e., the estimated working time of the UAV with the lightest task load. By calculating , the maximum difference in task completion time of all UAVs is obtained, and the goal of the entire formula is to minimize this difference. This means that during task allocation, the system will try to balance the task load of each UAV, avoiding the situation where some UAVs complete tasks prematurely and idle, and some UAVs have too heavy tasks, causing the overall inspection time to be prolonged. For example, in a large-scale city inspection task involving multiple UAVs, task allocation through this objective function can arrange the inspection range of each UAV according to the actual situation of each sub-area and the performance of the UAV, so that all UAVs can complete the task simultaneously as much as possible, thereby improving the inspection efficiency, fully utilizing the UAV resources, and reducing the overall inspection cost.

[0036] It is worth mentioning that: the embodiment describes a path optimization algorithm, taking the geometric center of the sub-region as the path node, using the improved A algorithm to search the path in the path search space constructed considering various constraints, and introducing a heuristic function containing distance and inspection complexity, and dynamically adjusting the path according to real-time data. The algorithm combines static environmental factors and dynamic data to achieve optimal planning of the path.

[0037] The technical effects achieved by the above embodiment include: the improved A algorithm combined with the heuristic function prioritizes paths with short distances and low inspection complexity during path search, reducing the flight of the unmanned aerial vehicle in complex areas and reducing energy consumption and risks. The constructed path search space considers various constraints to ensure that the planned path is practically feasible. The path is dynamically adjusted according to real-time data, allowing the unmanned aerial vehicle to avoid sudden obstacles, adverse weather areas or restricted areas in time, ensuring the smooth progress of the inspection task. The path optimization algorithm improves the scientificity, feasibility and dynamic adaptability of path planning, and improves the efficiency and safety of unmanned aerial vehicle inspection.

[0038] The traditional technical solution has the following technical problems: In the path planning algorithm, accurately calculating the node expansion cost is the key to finding the optimal path. If the actual cost is not accurately calculated, it may cause the path planning to deviate from the optimal solution, increasing the flight distance of the unmanned aerial vehicle and increasing energy consumption. Previous algorithms may oversimplify the calculation of the actual cost, not fully considering the actual situation of the unmanned aerial vehicle flight, affecting the accuracy of path planning. The present claim aims to accurately calculate the node expansion cost to improve the accuracy of path planning. Based on this, when using the improved algorithm to search the path, the node expansion cost function is: ; Wherein, is the actual cost from the starting node to the current node , calculated by the following formula: Wherein, , are the coordinates of the two adjacent path nodes, is the average flight speed of the unmanned aerial vehicle.

[0039] The formula explains the heuristic function used in the improved A* algorithm in the path optimization algorithm, which is used to evaluate the estimated cost from the current path node to the target path node . Its role is to guide the algorithm to more efficiently find the optimal path during path search. In the formula, represents the current node to the target node Euclidean distance, which is a common way to measure spatial distance, reflecting the direct distance from the current position to the target position. Euclidean distance is a fundamental consideration in path planning, and generally, the shorter the distance, the better the path. is the weight coefficient of the distance factor, used to adjust the importance of Euclidean distance in the heuristic function. In some high-time-demand and relatively simple environment inspection tasks, the value of can be appropriately increased to make the algorithm more inclined to choose a path with a short distance. represents the set of sub-regions allocated to the UAV The sum of the inspection complexity coefficients of all sub-regions in the set of sub-regions allocated to the UAV . As mentioned earlier, the inspection complexity coefficient considers factors such as the number of buildings, the proportion of tree coverage area, and the average terrain slope in the sub-region, reflecting the difficulty of sub-region inspection. Adding the complexity coefficients of all sub-regions can reflect the overall complexity faced by the UAV when executing the task path. is the weight coefficient of the complexity factor, used to adjust the influence of sub-region complexity in the heuristic function. In complex terrain or variable environment inspection areas, increasing can make the algorithm pay more attention to avoiding complex areas and choosing relatively simple and safe paths. and satisfy and are both greater than 0. By adjusting these two weight coefficients, the influence of distance and complexity factors on path selection can be flexibly balanced according to different inspection task requirements and environmental characteristics. For example, in a densely urban area, to ensure the safety of the UAV, it may be necessary to increase to preferentially select a path with low complexity; while in an open field area, to improve efficiency, it can be appropriate to increase to preferentially select a path with a short distance. This design enables the heuristic function to better adapt to diverse inspection scenarios, improving the scientificity and effectiveness of path planning. The formula explains that this formula is used to calculate the actual cost from the start node to the current node in the improved A* algorithm, which accurately reflects the cost consumed by the UAV on the flown path, providing accurate cost calculation basis for path search. In the formula, and represent the coordinates of two adjacent path nodes. By calculating , the straight-line distance between the two adjacent path nodes is obtained, which is based on the distance calculation formula of the plane rectangular coordinate system. Since the UAV flies along these path nodes in actual flight, the distance between adjacent nodes is accumulated, , the total flight distance from the start node to the current node . , which is an average speed value considering the performance of the UAV, environmental factors, etc. Dividing the total flight distance by the average flight speed, i.e. , the result is the time cost of the UAV flying from the start node to the current node , that is, the actual cost . In the path search process, is one of the important indicators for evaluating the pros and cons of the path. Together with the heuristic function , it constitutes the node expansion cost function . By accurately calculating , the actual consumption of the UAV on the flown path can be truly reflected. For example, in a path, if there are more turns or need to bypass obstacles, the distance between adjacent nodes will increase accordingly, resulting in increasing. When selecting a path, the algorithm will tend to choose a path with smaller. This makes the algorithm not only consider the estimated cost of the future path (reflected by ) when searching for the optimal path, but also fully consider the cost that has been consumed, so that the planned path is more in line with the actual flight requirements, improving the accuracy and rationality of path planning.

[0040] It is worth mentioning that: the embodiment further specifies the specific calculation method of the node expansion cost function in the improved A algorithm, wherein the actual cost from the start node to the current node is determined by calculating the ratio of the actual distance between adjacent path nodes to the average flight speed, combined with the heuristic function in the above embodiment, the node expansion cost is completely defined, providing accurate calculation basis for path search.

[0041] The technical effects achieved by the above embodiments include: by accurately calculating , the cost consumed by the UAV on the flown path can be truly reflected, combined with the heuristic function , the node expansion cost function more accurately evaluates the pros and cons of each node in the path search. In the search process of the improved A algorithm, based on the accurate cost function, the optimal path can be found more efficiently, avoiding unnecessary flight of the UAV, reducing flight distance and energy consumption, and improving the inspection efficiency. At the same time, accurate cost calculation also enhances the adaptability of path planning to different environments and task requirements, improving the overall quality of path planning.

[0042] The traditional technical solutions have the following technical problems: during the unmanned aerial vehicle inspection process, it is difficult to accurately assess the task execution risk in real time, and potential problems cannot be found in time and measures cannot be taken. For example, judging the state of the unmanned aerial vehicle by a single indicator may ignore other important factors, resulting in failure to warn of risks in advance. Lack of quantitative risk assessment criteria also makes it difficult to determine when to adjust the task, which can easily cause task delays or failures. Based on this, the flight state data of the unmanned aerial vehicle includes position coordinates , flight speed , flight attitude angle , and battery remaining capacity ; The task execution progress data includes the number of inspected sub-areas , and the amount of collected data ; The global intelligent platform calculates the task execution risk index of the unmanned aerial vehicle according to the flight state data and the task execution progress data by the following formula : ; Wherein, is the total capacity of the unmanned aerial vehicle battery, is the flight time, is the planned flight time, is the expected amount of collected data, , , is the weight coefficient, and , , , are all greater than 0; when the task execution risk index exceeds the preset threshold, the task reassignment and path adjustment mechanism is triggered.

[0043] This formula is used to calculate the unmanned aerial vehicle inspection task execution risk index , by comprehensively considering multiple key factors, the risk status during the unmanned aerial vehicle inspection task execution process is quantitatively evaluated, and the basis for timely taking risk response measures is provided. In the formula, represents the remaining capacity of the unmanned aerial vehicle battery, represents the total capacity of the unmanned aerial vehicle battery. reflects the remaining proportion of the battery capacity, and represents the consumption proportion of the battery capacity. is the weight coefficient of the battery capacity factor, which is used to adjust the influence degree of the battery capacity on the risk index. The battery capacity is one of the key factors for the unmanned aerial vehicle to successfully complete the task. If the capacity is consumed too quickly, the unmanned aerial vehicle may not be able to complete the remaining task or even fail to return, etc. By setting The importance of the battery power in the risk assessment can be highlighted according to actual needs. represents the flight time of the UAV, represents the planned flight time. The calculated is the absolute value of the deviation ratio of the flight time to the planned flight time. It reflects the time progress of the UAV task execution. If the actual flight time is much longer than the planned time, it may mean that difficulties are encountered in the task execution process, such as unreasonable path planning, environmental factors, etc. is a weight coefficient of the time factor, used to adjust the influence of the time progress on the risk index. In some inspection tasks with strict time requirements, The value of can be appropriately increased to more sensitively capture the risks in terms of time progress. represents the amount of data collected, represents the expected amount of data collected. The calculated is the absolute value of the deviation ratio of the collected data to the expected collected data. It reflects the task completion of the UAV in terms of data collection. If the actual collected data is much lower than expected, it may indicate that the device has failed or the task execution has problems. is a weight coefficient of the data collection factor, used to adjust the influence of the data collection on the risk index. In inspection tasks with data collection as the core target, The weight of should be correspondingly increased. , , satisfy , and are all greater than 0. By reasonably setting the three weight coefficients, the influence of the battery power, time progress, and data collection on the task execution risk can be comprehensively evaluated according to different inspection task characteristics and needs, and a comprehensive and accurate risk index When exceeds the preset threshold, the system can timely trigger the task reassignment and path adjustment mechanism, thereby effectively reducing the task execution risk and ensuring the smooth completion of the inspection task.

[0044] It is worth mentioning that: the embodiment determines the specific content of the UAV flight state data and the task execution progress data, and calculates the task execution risk index through the formula According to whether the risk index exceeds the threshold, the task reassignment and path adjustment mechanism is triggered. This scheme realizes the quantitative evaluation and dynamic management of the UAV inspection task execution risk.

[0045] The technical effects achieved by the above embodiment include: by comprehensively considering the battery power, flight time, and data collection amount, etc. to calculate the task execution risk index, the risk status of the UAV in the inspection task can be comprehensively and accurately evaluated. When the risk index exceeds the threshold value, timely triggering task reassignment and path adjustment can effectively avoid task failure caused by battery depletion, task timeout, or insufficient data collection, etc. This quantitative evaluation and dynamic management mechanism improves the reliability and stability of the UAV inspection task, ensures the smooth completion of the inspection task, and also improves the management ability of the global intelligent platform for the UAV inspection task.

[0046] The technical effects achieved by the above embodiment include: by comprehensively considering the battery power, flight time, and data collection amount, etc. to calculate the task execution risk index, the risk status of the UAV in the inspection task can be comprehensively and accurately evaluated. When the risk index exceeds the threshold value, timely triggering task reassignment and path adjustment can effectively avoid task failure caused by battery depletion, task timeout, or insufficient data collection, etc. This quantitative evaluation and dynamic management mechanism improves the reliability and stability of the UAV inspection task, ensures the smooth completion of the inspection task, and also improves the management ability of the global intelligent platform for the UAV inspection task. Based on this, the steps of the global intelligent platform reassigning the inspection task and adjusting the flight path are: reassigning the uncompleted task of the UAV that has failed or encountered an unexpected situation to the remaining normally operating UAV according to the task assignment algorithm; for each UAV after task reassignment, generating an optimal flight path according to the path optimization algorithm, and sending the new flight path to the corresponding UAV through the wireless communication module.

[0047] It is worth mentioning that: this embodiment details the specific operation of the global intelligent platform reassigning the inspection task and adjusting the flight path when the UAV fails or encounters unexpected situations. The uncompleted task of the failed UAV is reassigned to the remaining UAV according to the task assignment algorithm, and then a new path is generated for the UAV reassigned with the task according to the path optimization algorithm of the above embodiment and sent to the corresponding UAV, ensuring that the task can continue to be executed under abnormal conditions.

[0048] The traditional technical solution has the following technical problems: as the inspection tasks and the environment change continuously, the fixed task allocation and path optimization algorithm is difficult to continuously meet the demand of efficient inspection. If the algorithm cannot be optimized according to the actual situation, unreasonable task allocation and poor path planning may occur under new environmental or task requirements, resulting in a decrease in inspection efficiency. In addition, the lack of an update mechanism will make the algorithm gradually lag behind the development of technology and actual demand. The present application aims to solve the problems of poor adaptability and untimely updating of the algorithm. Based on this, it also includes: before the unmanned aerial vehicle performs the inspection task, the task allocation algorithm and the path optimization algorithm are trained and optimized by using historical inspection data and simulation data; the version of the task allocation algorithm and the path optimization algorithm in the global intelligent platform is updated regularly to adapt to different inspection task requirements and environmental changes.

[0049] It is worth mentioning that: the present application proposes that the task allocation algorithm and the path optimization algorithm are trained and optimized by using historical inspection data and simulation data before the unmanned aerial vehicle performs the inspection task, and the versions of the two algorithms in the global intelligent platform are updated regularly. Through data-driven and continuous optimization, the adaptability of the algorithm to different inspection tasks and environments is improved.

[0050] The technical effects achieved by the above embodiments include: training the algorithm by using historical inspection data and simulation data can make the algorithm learn the optimal strategy in different scenarios, improve the adaptability of the algorithm to complex environments and diversified tasks, and make the task allocation and path planning more accurate and efficient. Regular version update ensures that the algorithm can keep up with the changes in technology development and actual demand, continuously introduce new optimization methods and functions, and maintain the advancement of the algorithm. This continuous optimization and update mechanism improves the overall performance of the unmanned aerial vehicle inspection system under the global intelligent platform, prolongs the life cycle of the system, and provides protection for long-term stable and efficient inspection work.

[0051] The traditional technical solution has the following technical problems: in the unmanned aerial vehicle inspection system, the stability and security of data transmission are crucial. The traditional communication method may have the problems of slow transmission speed and unstable signal, resulting in data transmission delay or loss, affecting task execution. At the same time, if the data transmission is not encrypted, it is easy to be stolen or tampered with, revealing sensitive information and threatening system security. Based on this, the global intelligent platform and the unmanned aerial vehicle perform data transmission through a 5G communication network or a special wireless communication link, and the data transmission process adopts end-to-end encryption technology to ensure the security and integrity of data transmission.

[0052] It is worth mentioning that: the embodiment clearly shows that the global intelligent connection platform and the unmanned aerial vehicle transmit data through a 5G communication network or a dedicated wireless communication link, and adopts end-to-end encryption technology. The 5G communication network provides a high-speed and stable transmission channel, the dedicated wireless communication link serves as a backup to ensure transmission reliability, and the end-to-end encryption technology ensures the security and integrity of data during transmission.

[0053] The technical effects achieved by the above embodiment include: the combination of the 5G communication network and the dedicated wireless communication link ensures the rapid and stable transmission of data between the global intelligent connection platform and the unmanned aerial vehicle, so that the unmanned aerial vehicle can receive task instructions and send feedback data in a timely manner, ensuring the real-time and continuity of the inspection task. The end-to-end encryption technology effectively prevents data from being stolen or tampered with during transmission, protects the security and integrity of the inspection data, and avoids the leakage of sensitive information. This provides a solid communication guarantee for the reliable operation of the unmanned aerial vehicle inspection system, enhances the security and stability of the system, and improves the overall performance of the global intelligent connection platform.

[0054] In the above embodiments, the device elements are conventional device elements unless otherwise specified, and the connection methods and control methods are conventional connection methods and control methods unless otherwise specified.

[0055] The above has made a detailed description of the present application in combination with the embodiments, but those skilled in the art can understand that various specific parameters in the above embodiments can be changed to form multiple specific embodiments without departing from the purpose of the present application, which are within the common variation range of the present application, and will not be described one by one in detail.

Claims

1. A method for autonomous inspection and mission planning of drones under a global intelligent connection platform, characterized in that: include: Receive inspection task instructions from the global intelligent connection platform, which include the target inspection area, inspection accuracy requirements, and inspection time limit; Divide the target inspection area into multiple sub-areas, and calculate the inspection complexity coefficient of each sub-area based on the geographical environment, terrain height, and obstacle distribution of each sub-area; Based on the inspection complexity coefficient, inspection accuracy requirement and inspection time limit, the inspection task is assigned to multiple drones using a preset task allocation algorithm; For each sub-regional mission assigned to a drone, the optimal flight path is generated using a path optimization algorithm, taking into account the drone's endurance, flight speed, and payload weight. During the drone's inspection mission, the drone's flight status data and mission execution progress data are collected in real time and fed back to the global intelligent connection platform; When a drone malfunctions or encounters an emergency during a mission, the global intelligent connection platform reallocates inspection tasks and adjusts flight paths based on the status of the remaining drones and the mission execution status.

2. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 1 is characterized in that: The step of dividing the target inspection area into multiple sub-areas is specifically as follows: according to the shape and area of ​​the target inspection area, the target inspection area is divided into multiple square grid sub-areas of equal size using a grid division method; for each sub-area, its corresponding satellite map data, terrain height data and historical meteorological data are obtained to construct a sub-area environmental information database.

3. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 2 is characterized in that: The step of calculating the inspection complexity coefficient of each sub-area is specifically as follows: Determine the number of buildings in the sub-area based on the sub-area environmental information database , percentage of tree cover area , average terrain slope ; The inspection complexity coefficient is calculated using the following formula: : ； in, is the area of ​​the subregion, 、 、 is the weight coefficient, and , 、 、 Both are greater than 0.

4. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 1 is characterized in that: The preset task allocation algorithm specifically includes: Get the set of drones participating in the inspection mission , each drone The battery life is , the maximum flight speed is ; For each sub-region , according to its inspection complexity coefficient ,area And the preset inspection time standard per unit area , calculate the expected inspection time of the sub-area ; With the goal of minimizing the time difference in task completion of all drones, a task allocation model is established: ； in, Indicates the allocation to the drone The task allocation model is solved by genetic algorithm to obtain the optimal task allocation solution.

5. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 4 is characterized in that: The path optimization algorithm is specifically as follows: For each drone Assigned sub-region set , taking the geometric center of the sub-region as the path node; Considering the UAV's turning radius restrictions, no-fly zone restrictions, and flight altitude restrictions, a path search space is constructed; An improved algorithm is used to search for paths and introduce a heuristic function: ； in, is the current path node, is the target path node, is the Euclidean distance from the current node to the target node, 、 is the weight coefficient, and , 、 are greater than 0, For sub-region During the search process, the path is dynamically adjusted based on the real-time meteorological data and traffic control data.

6. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 5 is characterized in that: When the improved algorithm is used for path search, the node expansion cost function is: ； in, From the starting node to the current node The actual cost is calculated by the following formula: in, 、 are the coordinates of two adjacent path nodes, is the average flight speed of the UAV.

7. The method for autonomous inspection and mission planning of drones under the global intelligent connection platform according to claim 1 is characterized in that: The flight status data of the drone includes the position coordinates , flight speed , flight attitude angle , Remaining battery power ; The task execution progress data includes the number of inspected sub-areas , Amount of collected data ; The global intelligent connection platform calculates the UAV's mission execution risk index based on the flight status data and mission execution progress data using the following formula: : ； in, is the total power of the drone battery, is the flight time, To plan flight time, To estimate the amount of data to be collected, 、 、 is the weight coefficient, and , 、 、 are greater than 0; when the task execution risk index When the preset threshold is exceeded, the task reallocation and path adjustment mechanism is triggered.

8. The method for autonomous inspection and mission planning of drones under the global intelligent connection platform according to claim 5 is characterized in that: The steps of reallocating inspection tasks and adjusting flight paths on the global intelligent connection platform are specifically as follows: reallocating the unfinished tasks of drones that have malfunctioned or encountered emergencies to the remaining normally operating drones according to the task allocation algorithm; for each drone after the task is reallocated, regenerating the optimal flight path according to the path optimization algorithm, and sending the new flight path to the corresponding drone through the wireless communication module.

9. The method for autonomous inspection and mission planning of drones under the global intelligent connection platform according to claim 1 is characterized in that: Also includes: Before the UAV performs the inspection task, the task allocation algorithm and the path optimization algorithm are trained and optimized using historical inspection data and simulation data; The task allocation algorithm and path optimization algorithm in the global intelligent connection platform are regularly updated to adapt to different inspection task requirements and environmental changes.

10. The autonomous inspection and mission planning method for drones under the global intelligent connection platform according to claim 1 is characterized in that: Data is transmitted between the global intelligent connection platform and the drone through a 5G communication network or a dedicated wireless communication link. The data transmission process adopts end-to-end encryption technology to ensure the security and integrity of data transmission.

Images