Multi-unmanned aerial vehicle cooperative path planning and scheduling method and system

Abstract

This invention relates to a multi-UAV cooperative path planning and scheduling method and system. The method includes acquiring task requirements, determining multiple task areas requiring aerial surveying, coverage requirements, and UAV docking locations; constructing a coverage path with minimum turning cost for each task area based on coverage requirements; calculating the flight path cost of the UAVs based on their docking locations and coverage paths; and using a genetic algorithm to plan the areas and order of visits for each UAV based on the flight path cost, generating UAV flight trajectories. This method, by acquiring task requirements, generating UAV observation coverage paths with turn optimization within the task areas, and then using a genetic algorithm to generate optimized UAV scheduling schemes and flight trajectories based on coverage path costs, can minimize the total flight distance and balance the workload distribution among UAVs while ensuring high-quality and high-quantity completion of surveying requirements, thus shortening the data acquisition window for large-scale surveying.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method and system for collaborative path planning and scheduling of multiple UAVs. Background Technology

[0002] In large-area remote sensing scenarios, individual drones are inefficient and have limited endurance. Multi-drone swarms can cover different areas simultaneously through task splitting and parallel operations, significantly shortening the data acquisition time window for large areas. While multi-drone systems offer numerous advantages for remote sensing applications, building efficient collaborative scheduling and flight path planning mechanisms to fully realize their potential remains a major challenge. In complex scenarios with limited endurance, unclear target areas, or the need for multiple takeoffs and landings in multiple areas, inappropriate area allocation, access order, and flight path design can significantly increase ineffective flights and excessive turns, leading to increased energy consumption, shortened endurance, and consequently delayed data acquisition, impacting decision-making timeliness.

[0003] To improve deployment flexibility and continuous operation capabilities in remote sensing applications, existing technologies involve transforming ground vehicles into mobile platforms that integrate functions such as UAV transportation, rapid deployment, launch and recovery, airborne data processing, and energy replenishment—this is known as a vehicle-mounted UAV system. A vehicle-mounted multi-UAV system converts a ground vehicle into a mobile platform, integrating UAV transportation, rapid deployment, launch and recovery, data processing, and energy replenishment. This can solve the transportation, energy replenishment limitations, and computational bottlenecks faced by small and medium-sized UAVs in field operations.

[0004] However, most existing vehicle-mounted drone systems primarily support manual route planning for a single drone, making it difficult to adapt to the complex needs of multi-drone collaborative operations. Simply increasing the number of drones, without an effective coordination strategy, will lead to uneven task loads and inconsistent completion times among the drones, with overall efficiency limited by the "weakest link" effect.

[0005] Therefore, there is an urgent need to provide a collaborative path planning and scheduling scheme that can provide high-quality coverage of flight paths, avoid excessive turns by drones, and flexibly achieve high-quality coverage and time efficiency based on drone type. Summary of the Invention

[0006] This invention provides a multi-UAV cooperative path planning and scheduling method and system to overcome at least one of the above-mentioned technical problems in the prior art.

[0007] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions: In a first aspect, the present invention provides a multi-UAV cooperative path planning and scheduling method, comprising: Obtain mission requirements, determine multiple mission areas requiring aerial surveys, coverage requirements, and drone docking locations, where each mission area is independent of the others; Based on the coverage requirements, construct a coverage path with the minimum turning cost for each task area; Calculate the flight path cost of the drone based on the drone docking location and the coverage path; Based on the flight path cost, a genetic algorithm is used to plan the areas and order of visits for each UAV, generating the UAV flight trajectory.

[0008] In one possible implementation of the first aspect, constructing a coverage path with minimum turning cost for each task region according to the coverage requirement includes: Based on the coverage requirements, each task region is decomposed into an original topology map; Perform linear programming on the original topology graph to obtain an endpoint matching graph; The endpoint matching graph is matched using the Blossom V algorithm to obtain the matching results. The matching results are then projected onto the original topology graph to obtain the initial coverage path. The initial coverage path is optimized using local mixed integer programming to obtain the coverage path with the minimum turning cost.

[0009] In one possible implementation of the first aspect, the step of discretizing each task region into an original topology map according to the coverage requirement includes: Each task region is discretized into an original topology graph using equilateral triangles, wherein the distance between adjacent nodes in the original topology graph matches the coverage requirement.

[0010] In one possible implementation of the first aspect, performing linear programming on the original topology graph to obtain an endpoint matching graph includes: Introducing ternary pathway variables , representing path nodes Frequency of occurrence at each location, skip variables This indicates that if a path node is skipped, a value loss must be paid, and the value loss is... Using the following functions, linear programming is performed on the original topology graph to obtain a fractional solution; in, Represents the vertices of the original topology graph. Indicates the coverage value of the path node. Includes Euclidean distance and turning angle costs; The fractional solution is converted into multiple atomic bands, each atomic band representing a small imaginary movement at a path node with an adjacent edge orientation; Select the atomic bands with the best orientations in each of the path nodes as candidate atomic bands; For each candidate atomic band, construct corresponding endpoint nodes at both ends, and establish matching edges between any two endpoints to obtain an endpoint matching graph.

[0011] In one possible implementation of the first aspect, calculating the flight path cost of the UAV based on the UAV docking location and the coverage path includes: Based on the drone's docking location and the coverage path, the drone's load is calculated using the following formula: ; in, Indicates drone In the mission area The total length of the flight path required to complete coverage observations within the area. This represents the Euclidean distance between two locations. Indicates the drone docking position. Indicates the entry point. Indicates the exit point; The total path cost of a flight mission is calculated using the following formula: ; in, Indicates the number of drones. Indicates the drone's flight range; The load balance of the drone is calculated using the following formula: ; The total path cost and the load balance are weighted using a weighted summation mechanism to obtain the flight path cost.

[0012] In one possible implementation of the first aspect, generating a drone flight trajectory by using a genetic algorithm to plan the areas and order of visits for each drone based on the flight path cost includes: The task area is randomly assigned to each of the drones to obtain an initial population; The fitness of the path connection scheme for each individual in the initial population is calculated based on the flight path cost. Based on the fitness of the path connection scheme, the individuals in the population are selected using the NSGA-II framework to obtain the parent population individuals; The parent population individuals are processed using the multi-region path crossover operator of edge recombination to generate the offspring population individuals; According to the adaptive hybrid mutation mechanism, the offspring population individuals are mutated to obtain mutated offspring population individuals; The process is iterated until a preset number of iterations is reached, obtaining the planned areas and order of visits for each drone, and generating the drone flight trajectory.

[0013] In one possible implementation of the first aspect, prior to randomly assigning the task area to each of the drones to obtain an initial swarm, the method further includes: The UAV is encoded using a dual-chromosome encoding method, where the first layer represents the mission sequence and the second layer represents the UAV number.

[0014] In one possible implementation of the first aspect, the step of processing the parent population individuals according to the multi-region path crossover operator based on edge recombination to generate the offspring population individuals includes: Extract individuals from the parent population and assign them to the drone. Parental region and parent region and determine the drone In the parent region and parent region The set of task nodes that need to be accessed: ; For set Each node in Construct an adjacency list that includes regions from the parent region. and the parent region All edges constructed from the paths are added to the adjacency list, and the parent region is recorded. and the parent region The numbers of adjacent nodes, the distance between the edges, and an indication of whether the edge is in the parent region. and the parent region The mark that appears in both; offspring region Initialize as an empty sequence from the set Randomly select a starting node Add it to the offspring population and mark it as visited, so that the current node... ; Retrieve all neighboring nodes that the current node curr has not yet visited from the adjacency list; If a candidate neighbor node exists, then that candidate neighbor node is selected as the next node. ; If no candidate neighbor node exists, then from the set Randomly select an unvisited node as the next node. ; The next node Add the child region Mark as visited and update ; The retrieval process continues until the child region is reached. Length and the parent region The lengths are consistent, and the resulting child regions after intersection are obtained. ; offspring region Initialize to an empty sequence, then repeat the child region. The crossover step is used to obtain the crossover child region. ; The offspring region after all the drones are crossed and the child region Combine them to obtain offspring individuals.

[0015] In one possible implementation of the first aspect, the mutation processing of the offspring population individuals according to the adaptive hybrid mutation mechanism includes: According to the aforementioned adaptive hybrid mutation mechanism, with a probability of Select a combination of mutation operators The offspring population was subjected to mutation treatment; Wherein, the probability The probability is the current dynamic application probability of the combination of mutation operators. The dynamic adjustment method is as follows: Initialize an empty set of applied mutation operators. The combination of mutation operators Include Different mutation operators Each mutation operator Generate a probability ,if Then the mutation operator Add to the set of mutation operators The The mutation operator is The current dynamic application probability, the Initially, all the mutation operators mentioned above are used. The average value; With the set of mutation operators All of the aforementioned mutation operators The offspring population was subjected to mutation treatment; The fitness of the path connectivity schemes for each individual in the population before and after the mutation is compared, and the improvement score is calculated using the following formula. : ; in, Represents a very small positive number. This indicates the fitness of the modified path connection scheme; This indicates the fitness of the path connection scheme before the mutation; The improvement score Distributed equally to the set All of the aforementioned mutation operators For each of the mutation operators His contribution If recorded, >0, then each of the mutation operators The success counter increases by 1 ; Calculate each of the mutation operators In recent Average contribution in this application Each of the mutation operators is calculated using the following average contribution formula. New base probability: ; The base probability is normalized to obtain a new application probability.

[0016] Compared with the prior art, the present invention has at least the following beneficial effects: This invention provides a multi-UAV collaborative path planning and scheduling method. By acquiring task requirements, it generates UAV observation coverage paths with turn optimization within the task area. Then, based on the coverage path cost, it uses a genetic algorithm to generate an optimized UAV scheduling scheme and flight trajectory. This method can complete the mapping requirements with high quality and quantity while minimizing the total flight distance and balancing the workload distribution among UAVs, thus shortening the data acquisition window for large-scale mapping.

[0017] Secondly, the present invention provides a multi-UAV cooperative path planning and scheduling system, comprising: The acquisition module is used to acquire task requirements, determine multiple task areas that need aerial surveying, coverage requirements, and UAV docking locations, where each task area is independent of the others. The path generation module is used to construct a coverage path with the minimum turning cost for each task area according to the coverage requirements. The path cost calculation module is used to calculate the flight path cost of the UAV based on the UAV docking location and the coverage path. The path planning module is used to generate the drone flight trajectory by planning the areas and order to be visited by each drone based on the flight path cost using a genetic algorithm.

[0018] Thirdly, the present invention provides an electronic device comprising: at least one processor and at least one memory, wherein the memory stores computer-readable instructions; the computer-readable instructions are executed by one or more of the processors, causing the electronic device to implement the multi-UAV cooperative path planning and scheduling method as described in any implementation of the first aspect.

[0019] Fourthly, the present invention provides a storage medium having a computer-executable program stored thereon, the computer-executable program being used to cause a computer to execute the multi-UAV cooperative path planning and scheduling method as in any implementation of the first aspect.

[0020] Understandably, the beneficial effects achieved by the system of the second aspect, the electronic device of the third aspect, and the storage medium of the fourth aspect provided above can be referred to in light of the beneficial effects of the first aspect and any of its possible design embodiments, which will not be repeated here. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention; Figure 2 This is a flowchart of a multi-UAV cooperative path planning and scheduling method provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of a file outgoing leakage prevention system provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the present invention, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. The "or" in the present invention is merely a description of the relationship between the related objects, indicating that three relationships can exist. For example, A or B can represent: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Furthermore, in the description of the present invention, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items.

[0024] Furthermore, to facilitate a clear description of the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0025] In this embodiment of the invention, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this embodiment of the invention should not be construed as superior or more advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

[0026] In large-area remote sensing scenarios, individual drones are inefficient and have limited endurance. Multi-drone swarms can cover different areas simultaneously through task splitting and parallel operations, significantly shortening the data acquisition time window for large areas. While multi-drone systems offer numerous advantages for remote sensing applications, building efficient collaborative scheduling and flight path planning mechanisms to fully realize their potential remains a major challenge. In complex scenarios with limited endurance, unclear target areas, or the need for multiple takeoffs and landings in multiple areas, inappropriate area allocation, access order, and flight path design can significantly increase ineffective flights and excessive turns, leading to increased energy consumption, shortened endurance, and consequently delayed data acquisition, impacting decision-making timeliness.

[0027] To improve deployment flexibility and continuous operation capabilities in remote sensing applications, existing technologies involve transforming ground vehicles into mobile platforms that integrate functions such as UAV transportation, rapid deployment, launch and recovery, airborne data processing, and energy replenishment—this is known as a vehicle-mounted UAV system. A vehicle-mounted multi-UAV system converts a ground vehicle into a mobile platform, integrating UAV transportation, rapid deployment, launch and recovery, data processing, and energy replenishment. This can solve the transportation, energy replenishment limitations, and computational bottlenecks faced by small and medium-sized UAVs in field operations.

[0028] However, most existing vehicle-mounted drone systems primarily support manual route planning for a single drone, making it difficult to adapt to the complex needs of multi-drone collaborative operations. Simply increasing the number of drones, without an effective coordination strategy, will lead to uneven task loads and inconsistent completion times among the drones, with overall efficiency limited by the "weakest link" effect.

[0029] In view of this, on the one hand, embodiments of the present invention provide a multi-UAV cooperative path planning and scheduling method, including: obtaining task requirements, determining multiple task areas requiring aerial surveys, coverage requirements, and UAV docking positions, wherein each task area is independent of each other; constructing a coverage path with the minimum turning cost for each task area according to the coverage requirements; calculating the flight path cost of the UAVs according to the UAV docking positions and the coverage path; and using a genetic algorithm to plan the areas and order of visits for each UAV based on the flight path cost, thereby generating UAV flight trajectories.

[0030] This invention provides a multi-UAV collaborative path planning and scheduling method. By acquiring task requirements, it generates UAV observation coverage paths with turn optimization within the task area. Then, based on the coverage path cost, it uses a genetic algorithm to generate an optimized UAV scheduling scheme and flight trajectory. This method can complete the mapping requirements with high quality and quantity while minimizing the total flight distance and balancing the workload distribution among UAVs, thus shortening the data acquisition window for large-scale mapping.

[0031] In some embodiments, the multi-UAV cooperative path planning and scheduling method provided by the present invention can be executed by any electronic device 20 with data processing capabilities, such as a general-purpose computer, personal computer, laptop computer, switch or tablet computer, etc. The specific implementation of the electronic device 20 is not limited here.

[0032] Figure 1 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention is shown. The electronic device 20 includes a processor 210, a memory 220, and a communication interface 230.

[0033] Processor 210 may include one or more processing cores. Processor 210 connects to various parts within electronic device 20 using various interfaces and lines, and performs various functions and processes data of electronic device 20 by running or executing instructions, programs, code sets, or instruction sets stored in memory 220, and by calling data stored in memory 220. Optionally, processor 210 may be implemented using at least one of the following hardware forms: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA).

[0034] The memory 220 may include random access memory (RAI) or read-only memory (ROI). Optionally, the memory 220 may include non-transitory computer-readable storage ledger. The memory 220 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 220 may include a stored program area. The stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as data processing functions, data storage functions, and display push functions), and instructions for implementing the various method embodiments described above.

[0035] Communication interface 230 is used to communicate with other devices, equipment or communication networks, such as data storage devices, image processing devices or Ethernet, wireless access network (RAN), wireless local area network (WLAN), etc.

[0036] In terms of physical implementation, the aforementioned devices (such as processor 210, memory 220, and communication interface 230) can each be devices within the same device (such as a laptop computer). Alternatively, at least two of these devices can be located within the same device, i.e., as different devices within the same device, similar to the deployment of devices or components in a distributed system.

[0037] It is understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device 20. In other embodiments of the present invention, the electronic device 20 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0038] The following description, in conjunction with the accompanying drawings, illustrates a multi-UAV cooperative path planning and scheduling method provided by an embodiment of the present invention.

[0039] like Figure 2 As shown, this embodiment of the invention provides a multi-UAV cooperative path planning and scheduling method, which may include, but is not limited to: Step S1: Obtain mission requirements, determine multiple mission areas, coverage requirements, and UAV docking locations that need to be surveyed, where each mission area is independent of the others.

[0040] The task requirements in this embodiment of the invention may include, but are not limited to, task objectives and surveying requirements. Before starting the surveying task, several task areas that need to be surveyed by aerial photography are determined by the task objectives, as well as the location and boundaries of each task area. Based on the surveying requirements, the percentage of overlap between aerial images of adjacent flight paths, the percentage of overlap between aerial images of the same flight path, the flight altitude of the UAV, etc., are calculated, i.e., the coverage requirements.

[0041] In specific implementation, embodiments of the present invention may, but are not limited to, obtain the geographic center point of each region and the polygon sequence of the boundary of each region by using high-resolution satellite imagery and existing machine algorithms, and count the number of task regions.

[0042] In the specific implementation process, after determining the location of the mission area, the drone docking position is selected according to the actual situation on site. In this embodiment of the invention, a vehicle-mounted drone is used. The drone docking position is the same as the vehicle docking position. Usually, an open space with good communication signal is selected in the center of the mission area as the drone docking position.

[0043] Step S2: Based on the coverage requirements, construct a coverage path with the minimum turning cost for each task area.

[0044] In practical implementation, when a UAV needs to change direction during flight, it must decelerate to a certain speed and adjust its attitude. Additional thrust / power is needed to overcome centrifugal force and change direction safely. Fixed-wing UAVs, in particular, have minimum turning radius limitations based on speed, lift coefficient, and load coefficient. This invention, by constructing a coverage path with minimal turning cost for each mission area, reduces the number of sharp turns for the UAV, thereby lowering its power consumption.

[0045] In one feasible implementation, the step of constructing a coverage path with minimum turning cost for each task area according to the coverage requirements in this embodiment of the invention may include, but is not limited to: Based on the coverage requirements, each task region is decomposed into an original topology map; Perform linear programming on the original topology graph to obtain an endpoint matching graph; The endpoint matching graph is matched using the Blossom V algorithm to obtain the matching results. The matching results are then projected onto the original topology graph to obtain the initial coverage path. The initial coverage path is optimized using local mixed integer programming to obtain the coverage path with the minimum turning cost.

[0046] In specific implementation, the method described in this embodiment of the invention for dividing each task region into an original topology map according to the coverage requirements may include, but is not limited to: Each task region is discretized into an original topology graph using equilateral triangles, wherein the distance between adjacent nodes in the original topology graph matches the coverage requirement.

[0047] Specifically, in this embodiment of the invention, equilateral triangles are used to decompose each task region into an original topological graph. ; obtain a set of vertices and edge For each node Construct Voronoi cells on the mesh and use the area of ​​that cell as the coverage value for each node. This reflects the approximate coverage area when the path covers the node; where the edge The length is matched to the coverage requirements to meet the accuracy requirements of the surveying.

[0048] In specific implementation, the linear programming of the original topology graph to obtain the endpoint matching graph described in this embodiment of the invention may include, but is not limited to: Introducing ternary pathway variables , representing path nodes Frequency of occurrence at each location, skip variables This indicates that if a path node is skipped, a value loss must be paid, and the value loss is... Using the following functions, linear programming is performed on the original topology graph to obtain a fractional solution; in, Represents the vertices of the original topology graph. Indicates the coverage value of the path node. Includes Euclidean distance and turning angle costs; The fractional solution is converted into multiple atomic bands, each atomic band representing a small imaginary movement at a path node with an adjacent edge orientation; Select the atomic bands with the best orientations in each of the path nodes as candidate atomic bands; For each candidate atomic band, construct corresponding endpoint nodes at both ends, and establish matching edges between any two endpoints to obtain an endpoint matching graph.

[0049] In the specific implementation process, on any triangular mesh, each path node There can be multiple adjacent edge orientations. In this embodiment of the invention, the selection of the optimal several orientations of the atomic bands in each path node as candidate atomic bands can be, but is limited to, the following methods: First, enumerate the path nodes. All adjacent edges are taken as candidate orientations, and the path of the orientation is calculated. Central Order The path node is obtained by weighted summing the directions of its closest neighboring nodes. Orientation evaluation value. The optimal orientations for each node are retained as candidate atomic bands.

[0050] In specific implementation, the embodiment of the present invention describes constructing corresponding endpoint nodes at both ends of each candidate atomic band and establishing matching edges between any two endpoints to obtain an endpoint matching graph. This can be achieved, but is not limited to, using the following methods: If the two ends come from adjacent nodes and have the same geometric orientation, then the edge weight is equal to "half the distance from one end to the other + the turning cost"; if they are the two ends of the same atomic band, then the matching edge weight is set as the loss when that node is skipped. This results in an endpoint matching graph.

[0051] In specific implementation, the Blossom V algorithm used in this embodiment of the invention to match the endpoint matching graph, obtain matching results, and project the matching results onto the original topology graph to obtain the initial coverage path can be implemented, but is not limited to, the following methods: The minimum weight perfect match is solved using the Blossom-V algorithm in the endpoint matching graph, and the result is a set of pairs. Each pair of matches This either indicates connecting the two corresponding atomic band endpoints along the two ends to form a path segment, or it indicates skipping the node and paying a value loss. The matching result is then projected back to the original graph. This allows for the creation of several non-overlapping cycle covers, which form the initial coverage path. These cycle covers balance coverage value and turning costs at the global level, while avoiding excessive unnecessary backtracking.

[0052] In specific implementation, the method described in this embodiment of the invention for optimizing the initial coverage path using local mixed integer programming to obtain the coverage path with the minimum turning cost can be, but is not limited to, the following: Select the node with the "highest cost" Using it as the root, perform a breadth-first search (BFS) to construct a structure containing the previous... Subgraph of nodes In this subgraph, the corresponding path and skip variables are set as binary (0 / 1) variable variables, while the remaining external variables are fixed as constants rounded up from the fractional solution. After solving the entire subgraph, if the objective value decreases and the local solution can connect with the external solution, the global solution is updated. Redundant turns and unnecessary skipping of nodes are reduced through repeated iterations. A Prize-Collecting Steiner Tree (PCST) model is introduced, where the edge weight between two closed loops represents their shortest connection cost. Solving the PCST yields a "value-maximizing" connected subtree. Then, only the selected closed loops in the PCST are merged, while the rest are discarded. For geometrically adjacent closed loops with low connection costs, Dijkstra's algorithm is used to calculate the shortest path with turning costs, and these are merged into a larger closed loop. Finally, a global-scale local MIP optimization is performed on this large closed loop to further eliminate redundancy and overlapping coverage generated during merging. The covered path with the minimum turning cost is ultimately obtained.

[0053] Step S3: Calculate the flight path cost of the drone based on the drone docking location and the coverage path.

[0054] In one feasible implementation, the calculation of the drone's flight path cost based on the drone's docking location and the coverage path in this embodiment of the invention includes: Based on the drone's docking location and the coverage path, the drone's load is calculated using the following formula: ; in, Indicates drone In the mission area The total length of the flight path required to complete coverage observations within the area. This represents the Euclidean distance between two locations. Indicates the drone docking position. Indicates the entry point. Indicates the exit point; The total path cost of a flight mission is calculated using the following formula: ; in, Indicates the number of drones. Indicates the drone's flight range; The load balance of the drone is calculated using the following formula: ; The total path cost and the load balance are weighted using a weighted summation mechanism to obtain the flight path cost.

[0055] Step S4: Based on the flight path cost, use a genetic algorithm to plan the areas and order of visits for each UAV and generate the UAV flight trajectory.

[0056] In one feasible implementation, the step of generating a drone flight trajectory by using a genetic algorithm to plan the areas and order of visits for each drone based on the flight path cost in this embodiment of the invention may include, but is not limited to: The task area is randomly assigned to each of the drones to obtain an initial population; The fitness of the path connection scheme for each individual in the initial population is calculated based on the flight path cost. Based on the fitness of the path connection scheme, the individuals in the population are selected using the NSGA-II framework to obtain the parent population individuals; The parent population individuals are processed using the multi-region path crossover operator of edge recombination to generate the offspring population individuals; According to the adaptive hybrid mutation mechanism, the offspring population individuals are mutated to obtain mutated offspring population individuals; The process is iterated until a preset number of iterations is reached, obtaining the planned areas and order of visits for each drone, and generating the drone flight trajectory.

[0057] It should be noted that the iterative loop in the embodiments of the present invention refers to, after obtaining the mutant offspring population individuals, calculating the fitness of the path connection scheme of the mutant offspring population individuals based on the flight path cost, then using the NSGA-II framework to select the population, and then crossover and mutation, until the preset number of iterations is reached, and finally determining the area and order to be visited by each UAV.

[0058] In specific implementation, after determining the planned areas and order of each UAV's visits, the embodiments of the present invention can, but are not limited to, use Dubins paths to connect the entry and exit points of each area, generating a smooth flight trajectory that conforms to the dynamic constraints of the UAV.

[0059] In the specific implementation process, before randomly assigning the task area to each UAV to obtain the initial population, this embodiment of the invention further includes: The UAV is encoded using a dual-chromosome encoding method, where the first layer represents the mission sequence and the second layer represents the UAV number.

[0060] Specifically, in the first layer, each drone is assigned a list, with the head and tail representing the drone's docking position (vehicle location). After task initialization, the middle part of the list is filled with area access tasks. The vehicle index is always at the beginning or end of each drone's task allocation list and its position remains unchanged. Using dual-chromosome encoding to encode drones overcomes the limitations of traditional methods on the number of tasks, supporting dynamic task allocation and load balancing.

[0061] In specific implementation, the process of processing the parent population individuals based on the multi-region path crossover operator of edge recombination to generate the offspring population individuals in this embodiment of the invention may include, but is not limited to: Extract individuals from the parent population and assign them to the drone. Parental region and parent region and determine the drone In the parent region and the parent region The set of task nodes that need to be accessed: ; For set Each node in Construct an adjacency list that includes regions from the parent region. and the parent region All edges constructed from the paths are added to the adjacency list, and the parent region is recorded. and the parent region The numbers of adjacent nodes, the distance between the edges, and an indication of whether the edge is in the parent region. and the parent region The mark that appears in both; offspring region Initialize as an empty sequence from the set Randomly select a starting node Add it to the offspring population and mark it as visited, making the current node... ; Retrieve all neighboring nodes that the current node curr has not yet visited from the adjacency list; If a candidate neighbor node exists, then that candidate neighbor node is selected as the next node. ; If no candidate neighbor node exists, then from the set Randomly select an unvisited node as the next node. ; The next node Add the child region Mark as visited and update ; The retrieval process continues until the child region is reached. Length and the parent region The lengths are consistent, and the resulting child regions after intersection are obtained. ; offspring region Initialize to an empty sequence and repeat the above child region. The crossover step is used to obtain the crossover child region. ; The offspring region after all the drones are crossed and the child region Combine them to obtain offspring individuals.

[0062] In specific implementation, the mutation processing of the offspring population individuals according to the adaptive hybrid mutation mechanism described in this embodiment of the invention includes: According to the aforementioned adaptive hybrid mutation mechanism, with a probability of Select a combination of mutation operators The offspring population was subjected to mutation treatment; Wherein, the probability The probability is the current dynamic application probability of the combination of mutation operators. The dynamic adjustment method is as follows: Initialize an empty set of applied mutation operators. The combination of mutation operators Include Different mutation operators Each mutation operator Generate a probability ,if Then the mutation operator Add to the set of mutation operators The The mutation operator is The current dynamic application probability, the Initially, all the mutation operators mentioned above are used. The average value; With the set of mutation operators All of the aforementioned mutation operators The offspring population was subjected to mutation treatment; The fitness of the path connectivity schemes for each individual in the population before and after the mutation is compared, and the improvement score is calculated using the following formula. : ; in, Represents a very small positive number. This indicates the fitness of the modified path connection scheme; This indicates the fitness of the path connection scheme before the mutation; The improvement score Distributed equally to the set All of the aforementioned mutation operators For each of the mutation operators His contribution If recorded, >0, then each of the mutation operators The success counter increases by 1 ; Calculate each of the mutation operators In recent Average contribution in this application Each of the mutation operators is calculated using the following average contribution formula. New base probability: ; The base probability is normalized to obtain a new application probability.

[0063] This invention provides an application experiment of the above-mentioned multi-UAV cooperative path planning and scheduling method. In a disaster-stricken building survey in a certain area, three UAVs were used to survey 13 damaged building areas. The three UAVs undertook 20.9%, 35.4%, and 43.7% of the aerial survey tasks, respectively. In a village cluster area survey in a certain area, six UAVs surveyed 42 areas, and their respective tasks were 17%, 17%, 16%, 16%, 17%, and 17%.

[0064] The multi-UAV cooperative path planning and scheduling method provided by the present invention obtains task requirements, generates UAV observation coverage paths with turn optimization within the task area, and then uses a genetic algorithm to generate optimized UAV scheduling schemes and flight trajectories based on coverage path costs. This method can minimize the total flight distance and balance the workload distribution among UAVs while ensuring the quality and quantity of mapping requirements, thus shortening the data acquisition window for large-scale mapping.

[0065] Based on the multi-UAV cooperative path planning and scheduling method provided in the first aspect, embodiments of the present invention provide a multi-UAV cooperative path planning and scheduling system, such as... Figure 3 As shown, the multi-UAV cooperative path planning and scheduling system includes: The acquisition module 110 is used to acquire task requirements, determine multiple task areas, coverage requirements and UAV docking locations that need to be surveyed by air, wherein each task area is independent of the others. Path generation module 120 is used to construct a coverage path with the minimum turning cost for each task area according to the coverage requirements; The path cost calculation module 130 is used to calculate the flight path cost of the UAV based on the UAV docking location and the coverage path. The path planning module 140 is used to generate a drone flight trajectory by using a genetic algorithm to plan the areas and order to be visited by each drone based on the flight path cost.

[0066] Based on the multi-UAV cooperative path planning and scheduling method provided in the first aspect, this embodiment of the invention also provides a storage medium storing a computer-executable program. The computer-executable program is used to cause a computer to execute the multi-UAV cooperative path planning and scheduling method as described in any implementation of the first aspect. Explanations of the relevant content and descriptions of the beneficial effects of any of the computer-readable storage media provided above can be found in the corresponding embodiments described above, and will not be repeated here.

[0067] Those skilled in the art will understand that the program for implementing all or part of the steps of the above embodiments, which can be executed by a program instructing related hardware, can be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a random access memory, etc. The processing unit or processor mentioned above can be a central processing unit, a general-purpose processor, an application-specific integrated circuit (ASIC), a microprocessor (DSP), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0068] This invention also provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform any of the methods described in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0069] It should be noted that the devices for storing computer instructions or computer programs provided in the embodiments of the present invention, such as, but not limited to, the aforementioned memory, computer-readable storage medium, and communication chip, are all non-transitory. Those skilled in the art should recognize that the functions described in the embodiments of the present invention in one or more of the above examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable storage medium or transmitted as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium accessible to general-purpose or special-purpose computers.

[0070] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A multi-UAV cooperative path planning and scheduling method, characterized in that, include: Obtain mission requirements, determine multiple mission areas requiring aerial surveys, coverage requirements, and drone docking locations, where each mission area is independent of the others; Based on the coverage requirements, construct a coverage path with minimum turning cost for each task area, including: According to the coverage requirements, each task region is discretized into an original topology graph using equilateral triangles, wherein the distance between adjacent nodes in the original topology graph matches the coverage requirements; Introducing ternary path variables , representing the frequency of occurrence at path nodes , skipping variables representing the value loss paid if a path node is skipped, the loss value being , linear programming of the original topology graph using the following functions to obtain a fractional solution; wherein, denotes a set of vertices of the original topology graph, denotes a coverage value of a path node, contains a Euclidean distance and a turning angle cost; The fractional solution is converted into multiple atomic bands, each atomic band representing a small imaginary movement at a path node with an adjacent edge orientation; Select the atomic bands with the best orientations in each of the path nodes as candidate atomic bands; For each candidate atomic band, construct corresponding endpoint nodes at both ends, and establish matching edges between any two endpoints to obtain an endpoint matching graph; The endpoint matching graph is matched using the Blossom V algorithm to obtain the matching results. The matching results are then projected onto the original topology graph to obtain the initial coverage path. The initial coverage path is optimized using local mixed integer programming to obtain the coverage path with the minimum turning cost; Based on the drone's docking location and the coverage path, the drone's load is calculated using the following formula: ； wherein, denotes a UAV in a mission area total length of the flight path required to complete the coverage observation, denotes the Euclidean distance between two locations, denotes a UAV parking position, denotes an entry point, denotes an exit point; The total path cost of a flight mission is calculated using the following formula: ； wherein, represents the number of drones, represents the drone range; The load balance of the drone is calculated using the following formula: ； The total path cost and the load balance are weighted using a weighted summation mechanism to obtain the flight path cost; Based on the flight path cost, a genetic algorithm is used to plan the areas and order of visits for each UAV, generating the UAV flight trajectory.

2. The method of claim 1, wherein, The step of using a genetic algorithm to plan the areas and order of visits for each UAV based on the flight path cost, and generating the UAV flight trajectory, includes: The task area is randomly assigned to each of the drones to obtain an initial population; The fitness of the path connection scheme for each individual in the initial population is calculated based on the flight path cost. Based on the fitness of the path connection scheme, the individuals in the population are selected using the NSGA-II framework to obtain the parent population individuals; The parent population individuals are processed using the multi-region path crossover operator of edge recombination to generate the offspring population individuals; According to the adaptive hybrid mutation mechanism, the offspring population individuals are mutated to obtain mutated offspring population individuals; The process is iterated until a preset number of iterations is reached, obtaining the planned areas and order of visits for each drone, and generating the drone flight trajectory.

3. The method of claim 2, wherein, Before randomly assigning the task area to each of the drones to obtain an initial population, the method further includes: The UAV is encoded using a dual-chromosome encoding method, where the first layer represents the mission sequence and the second layer represents the UAV number.

4. The multi-UAV cooperative path planning and scheduling method according to claim 3, characterized in that, The process of using the multi-region path crossover operator with edge recombination to process the parent population individuals and generate the offspring population individuals includes: Extract individuals from the parent population and assign them to the drone. Parental region and parent region and determine the drone In the parent region and parent region The set of task nodes that need to be accessed: ； For set Each node in Construct an adjacency list that includes regions from the parent region. and the parent region All edges constructed from the paths are added to the adjacency list, and the parent region is recorded. and the parent region The numbers of adjacent nodes, the distance between the edges, and an indication of whether the edge is in the parent region. and the parent region The mark that appears in both; offspring region Initialize as an empty sequence from the set Randomly select a starting node Add it to the offspring population and mark it as visited, so that the current node... ; Retrieve all neighboring nodes that the current node curr has not yet visited from the adjacency list; If a candidate neighbor node exists, then that candidate neighbor node is selected as the next node. ; If no candidate neighbor node exists, then from the set Randomly select an unvisited node as the next node. ; The next node Add the child region Mark as visited and update ; The retrieval process continues until the child region is reached. Length and the parent region The lengths are consistent, and the resulting child regions after intersection are obtained. ; offspring region Initialize to an empty sequence, then repeat the child region. The crossover step is used to obtain the crossover child region. ; The offspring region after all the drones are crossed and the child region Combine them to obtain offspring individuals.

5. A multi-UAV cooperative path planning and scheduling method according to claim 2, characterized in that, The mutation process performed on the offspring population individuals according to the adaptive hybrid mutation mechanism includes: According to the aforementioned adaptive hybrid mutation mechanism, with a probability of Select a combination of mutation operators The offspring population was subjected to mutation treatment; Wherein, the probability The probability is the current dynamic application probability of the combination of mutation operators. The dynamic adjustment method is as follows: Initialize an empty set of applied mutation operators. The combination of mutation operators Include Different mutation operators Each mutation operator Generate a probability ,if Then the mutation operator Add to the set of mutation operators The The mutation operator is The current dynamic application probability, the Initially, all the mutation operators mentioned above are used. The average value; With the set of mutation operators All of the aforementioned mutation operators The offspring population was subjected to mutation treatment; The fitness of the path connectivity schemes for each individual in the population before and after the mutation is compared, and the improvement score is calculated using the following formula. : ； in, Represents a very small positive number. This indicates the fitness of the modified path connection scheme; This indicates the fitness of the path connection scheme before the mutation; The improvement score Distributed equally to the set All of the aforementioned mutation operators For each of the mutation operators His contribution If recorded, >0, then each of the mutation operators The success counter increases by 1 ; Calculate each of the mutation operators In recent Average contribution in this application Each of the mutation operators is calculated using the following average contribution formula. New base probability: ； The base probability is normalized to obtain a new application probability.

6. A multi-UAV cooperative path planning and scheduling system, characterized in that, include: The acquisition module is used to acquire task requirements, determine multiple task areas that need aerial surveying, coverage requirements, and UAV docking locations, where each task area is independent of the others. The path generation module is used to construct a coverage path with the minimum turning cost for each task area according to the coverage requirements, including: According to the coverage requirements, each task region is discretized into an original topology graph using equilateral triangles, wherein the distance between adjacent nodes in the original topology graph matches the coverage requirements; Introducing ternary pathway variables , representing path nodes Frequency of occurrence at each location, skip variables This indicates that if a path node is skipped, a value loss must be paid, and the value loss is... Using the following functions, linear programming is performed on the original topology graph to obtain a fractional solution; in, This represents the set of vertices in the original topological graph. Indicates the coverage value of the path node. Includes Euclidean distance and turning angle costs; The fractional solution is converted into multiple atomic bands, each atomic band representing a small imaginary movement at a path node with an adjacent edge orientation; Select the atomic bands with the best orientations in each of the path nodes as candidate atomic bands; For each candidate atomic band, construct corresponding endpoint nodes at both ends, and establish matching edges between any two endpoints to obtain an endpoint matching graph; The endpoint matching graph is matched using the Blossom V algorithm to obtain the matching results. The matching results are then projected onto the original topology graph to obtain the initial coverage path. The initial coverage path is optimized using local mixed integer programming to obtain the coverage path with the minimum turning cost; The path cost calculation module is used to calculate the drone's load based on the drone's docking location and the coverage path using the following formula: ； in, Indicates drone In the mission area The total length of the flight path required to complete coverage observations within the area. This represents the Euclidean distance between two locations. Indicates the drone docking position. Indicates the entry point. Indicates the exit point; The total path cost of a flight mission is calculated using the following formula: ； in, Indicates the number of drones. Indicates drone The driving range; The load balance of the drone is calculated using the following formula: ； The total path cost and the load balance are weighted using a weighted summation mechanism to obtain the flight path cost; The path planning module is used to generate the drone flight trajectory by planning the areas and order to be visited by each drone based on the flight path cost using a genetic algorithm.

Images