Unmanned aerial vehicle cluster management system and method based on intelligent algorithm

Abstract

The invention relates to the technical field of unmanned aerial vehicle path planning, in particular to an unmanned aerial vehicle cluster management system and method based on an intelligent algorithm, and the method comprises the steps: obtaining the operation data of each unmanned aerial vehicle and the attribute data of each flight task in each dimension; constructing a constraint condition and a path optimization objective function in combination with the flight path of the unmanned aerial vehicle according to the distribution condition of the load in the operation data of each unmanned aerial vehicle and the attribute data of the time dimension; and constructing a path node graph according to the grid map, updating pheromones according to Tent chaotic mapping and an ant colony algorithm, and combining the path optimization objective function to obtain an optimal path planning scheme of the unmanned aerial vehicle cluster. According to the invention, through joint optimization of task allocation and path planning, the cooperative efficiency of the unmanned aerial vehicle cluster is improved. In combination with an intelligent algorithm, on the basis of considering obstacles, flight area limitation and task priority, real-time dynamic adjustment of path planning is realized, and efficient and safe flight of the unmanned aerial vehicle in a complex environment is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to an intelligent algorithm-based UAV cluster management system and method. Background Art

[0002] With the rapid development of drone technology, drone swarm collaborative operations have been widely used in many fields. Effective path planning and task allocation are crucial to improving the execution efficiency and safety of drone swarms. Currently, most existing algorithms are limited to single drone path planning, or perform well in simple environments, but are inefficient and prone to collisions in complex environments. In addition, the traditional method is to solve and calculate path planning and task allocation separately, and the effectiveness of path planning and task allocation is affected by many factors, such as task priority, path length, and the dynamic characteristics of drones. Therefore, the existing drone swarm path planning methods do not consider the coupling relationship between task allocation and path optimization, and cannot meet the needs of actual engineering. Summary of the invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a drone cluster management system and method based on intelligent algorithm. The technical solutions adopted are as follows:

[0004] In a first aspect, the present invention provides a method for managing a drone cluster based on an intelligent algorithm, comprising:

[0005] Obtain the operating data of each drone in the drone cluster and the attribute data of each flight mission in various dimensions;

[0006] According to the load distribution and time dimension attribute data in the operation data of each drone in the drone cluster, combined with the flight path of the drone, the constraint conditions and path optimization objective function are constructed;

[0007] A path node graph is constructed according to the grid map, pheromones are updated according to the Tent chaotic map and ant colony algorithm, and the optimal path planning scheme of the UAV cluster is obtained by combining the path optimization objective function.

[0008] Preferably, the operating data of the UAV includes the maximum load data and maximum cruising range of the UAV; the attribute data of the flight mission in various dimensions include the priority of the flight mission, the required material weight, the start time and end time of the flight mission, and the flight distance between different flight missions.

[0009] Preferably, the constraints specifically include:

[0010] The weight of materials required for the corresponding flight mission performed by the drone shall not exceed the maximum load data of each drone; the mission time performed by the drone shall be between the start time and end time of the corresponding flight mission; the flight mission performed by the drone shall be continuous; the total flight distance of the drone shall not exceed the maximum endurance distance.

[0011] Preferably, the path optimization objective function specifically includes:

[0012]

[0013] Wherein, F represents the path optimization objective function, λ1, λ2 and λ3 all represent preset adjustment coefficients, n represents the total number of drones, m represents the total number of drone flight missions, i represents the i-th drone, j represents the j-th flight mission, and k represents the k-th flight mission;

[0014] d jk represents the flight distance between the jth flight mission and the kth flight mission of the UAV, p j represents the priority of the jth flight mission, x ij represents the first decision variable between the i-th UAV and the j-th flight mission, x ij ∈{0,1},y ijk represents the second decision variable between the i-th UAV and the j-th flight mission and the k-th flight mission, y ijk ∈{0,1};C i represents the maximum load data of the i-th UAV, l j Represents the required material weight of the j-th flight mission.

[0015] Preferably, the step of constructing a path node graph according to the grid map specifically includes:

[0016] A grid map is constructed based on the working environment of the drone cluster, wherein the element value of the obstacle in the grid map is 1, the element value of the task target is 2, and the element value of the free grid is 0;

[0017] According to the grid number of each grid in the grid map and the horizontal coordinate scale and vertical coordinate scale of the grid, the label of each grid in the grid map is obtained;

[0018] In the grid map, a square composed of four adjacent grids is the basic unit. If only one grid in the basic unit is marked as an obstacle, the center point of the basic unit corresponding to the square is used as the key feature point, and the key feature point is used as the node of path planning to construct a path node graph.

[0019] Preferably, obtaining the label of each grid in the grid map according to the grid number of each grid in the grid map and the horizontal coordinate scale and the vertical coordinate scale of the grid specifically includes:

[0020]

[0021] Among them, x v and v They represent the labels corresponding to the grid number v, mod is the remaining operator, N x Indicates the horizontal scale of the grid, N y Represents the vertical scale of the grid, cell is the round-up function, and v is the grid number.

[0022] Preferably, the formula of the Tent chaotic mapping is specifically:

[0023]

[0024] Among them, X(t+1) represents the chaotic sequence value at time t+1, X(t) represents the chaotic sequence value at time t, X(t)∈(0,1), and t represents time t.

[0025] Preferably, the updating of pheromones according to Tent chaotic mapping and ant colony algorithm specifically includes:

[0026]

[0027] τ ij (t+n)=(1-ρ)·τ ij (t)+Δτ ij (t)

[0028]

[0029] Among them, τ ij (t) is the pheromone concentration between node i and node j at time t; ρ is the information volatility factor; α is the information heuristic factor; β is the expected heuristic factor; H(d ij ,Cost(U i ,T j )) is the comprehensive cost of UAV formation task allocation and path planning; Cost(U i ,T j ) is the cost of the ant colony selecting task i according to the difficulty of task execution, U i represents the i-th drone, T j represents the jth flight mission; d ij represents the distance between the i-th UAV and the j-th flight mission of the UAV, N m lrepresents the set of neighboring nodes of ant m at node l, Δτ ij m (t) represents the local increase of ant m’s pheromone on edge (i, j), Δτ ij (t) represents the total increase of pheromone of all ants on edge (i, j), p ij m represents the probability of ant m moving from node i to node j, τ ij (t+n) represents the pheromone concentration between node i and node j at time t+n.

[0030] In a second aspect, the present invention provides a drone cluster management system based on an intelligent algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the computer program is executed by the processor, the steps of a drone cluster management method based on an intelligent algorithm are implemented.

[0031] The embodiments of the present invention have at least the following beneficial effects:

[0032] 1. Improve task execution efficiency: By setting a joint optimization objective function, the task allocation efficiency, path length and task priority can be comprehensively considered to maximize the overall execution efficiency and ensure that the drone can complete the assigned tasks quickly and efficiently.

[0033] 2. Ensure path safety: The improved path planning method ensures that there is no intersection between the generated paths, effectively avoiding collisions between drones, reducing operational risks and improving flight safety.

[0034] 3. Achieve load balancing: During the task allocation process, strive to maintain load balance among the drones so that each drone can participate in the task appropriately, prevent some drones from being overloaded while other drones are idle, and improve overall operational efficiency.

[0035] 4. Improve path selection diversity: The improved ant colony algorithm increases the flexibility and diversity of path selection through uneven pheromone distribution and bidirectional parallel path search, thereby reducing the probability of deadlock and path selection dilemma.

[0036] 5. Economic benefits: By improving the overall operational efficiency and safety of drone clusters, operating costs can be reduced and production efficiency can be improved, thus bringing significant economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 It is a flowchart of the steps of a drone cluster management method based on an intelligent algorithm provided by the present invention;

[0039] Figure 2 It is a design schematic diagram based on the decimal formation ID provided by the present invention;

[0040] Figure 3 It is a flowchart of the improved ant colony algorithm provided by the present invention;

[0041] Figure 4 It is a flowchart of the steps of the method for obtaining the path node graph provided by the present invention. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a drone cluster management system and method based on an intelligent algorithm proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of a drone cluster management system and method based on an intelligent algorithm provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of a method for managing a drone cluster based on an intelligent algorithm provided by an embodiment of the present invention, the method comprising the following steps:

[0046] Step 1: Obtain the operating data of each drone in the drone cluster and the attribute data of each flight mission in various dimensions.

[0047] Specifically, the operation data of the UAV includes the maximum load data and maximum cruising distance of the UAV; the attribute data of the flight mission in various dimensions include the priority of the flight mission, the required material weight, the start time and end time of the flight mission, and the flight distance between different flight missions.

[0048] Among them, the drone cluster set can be expressed as U = {U1,U2,…,U n}, n represents the total number of drones, and the task set can be expressed as T = {T1, T2, …, T m}, m represents the total number of flight missions.

[0049] It should be noted that the marking and extraction of mission targets helps the drone identify its mission area and provides data support for path planning. Each mission target contains additional attributes, which are crucial for task scheduling and drone resource allocation. In this embodiment, the priority of the flight mission is determined as follows:

[0050] First, the priority based on time: tasks with earlier start time and shorter time window have higher priority. The calculation formula for the first priority of tasks can be expressed as:

[0051]

[0052] Among them, p j time represents the first priority of the jth task, T end,j and T start,j are the end time and start time of the jth task respectively, ∈ is the minimum value to prevent the denominator from being zero, and its value can be 0.1.

[0053] Based on task importance: The priority of a task is assigned a basic weight based on the nature of the task itself (such as the urgency or strategic importance of the task). Urgent or critical tasks have a higher priority weight. The weight can be pre-set by the task attribute data. The formula is as follows:

[0054] P j importance =W j

[0055] Among them, P j importance represents the second priority of the jth task, W j is the importance weight of the j-th task.

[0056] Based on the weight of required materials: Tasks that require heavy materials may affect the endurance of the drone, so the priority is appropriately reduced. The specific calculation formula is:

[0057]

[0058] Among them, P j weight represents the third priority of the jth task, l j is the weight of materials required for the jth task.

[0059] Finally, the comprehensive priority is calculated by weighting the priorities of factors such as time, task importance, and material weight to obtain the comprehensive priority of the task, which can be expressed as:

[0060] P j =γ1·P j time +γ2·P j importance +γ3·P j weight

[0061] Among them, P j It represents the comprehensive priority of the jth task. γ1, γ2, and γ3 are weight coefficients, and their values ​​are adjusted according to the specific application scenario.

[0062] Through the above method, it is ensured that the priority determination can comprehensively reflect the time urgency, importance and resource consumption of the task, and provide a scientific basis for subsequent task scheduling and path planning.

[0063] Step 2: Based on the load distribution in the operation data of each drone in the drone cluster and the attribute data in the time dimension, combined with the drone flight path, construct constraint conditions and path optimization objective function.

[0064] In engineering construction scenarios, such as construction monitoring, material distribution, and equipment inspection, drone swarms need to perform a variety of tasks. To ensure that drones can complete these tasks efficiently within a limited time, a joint optimization model for task allocation and path planning needs to be established. This model comprehensively considers constraints such as time window, flight distance, task priority, and drone load, and realizes the coordinated optimization of path planning and task allocation.

[0065] By constructing constraints, the effectiveness of task allocation and path planning is ensured. The constraints include: the weight of materials required for the corresponding flight mission performed by the drone is not greater than the maximum load data of each drone; the mission time performed by the drone is between the start time and end time of the corresponding flight mission; the flight mission performed by the drone is continuous; the total flight distance of the drone is not greater than the maximum endurance distance.

[0066] Among them, the flight mission performed by the UAV has continuity, which means that if the UAV starts from the mission point T j Fly to mission point T k, then it must perform the flight mission T j .

[0067] By constructing a path optimization objective function, the coordinated optimization of path planning and task allocation is achieved. The path optimization objective function specifically includes:

[0068]

[0069] Wherein, F represents the path optimization objective function, λ1, λ2 and λ3 represent preset adjustment coefficients, n represents the total number of UAVs, m represents the total number of UAV flight missions, i represents the i-th UAV, j represents the j-th flight mission, and k represents the k-th flight mission; d jk represents the flight distance between the jth flight mission and the kth flight mission of the UAV, p j represents the priority of the jth flight mission, x ij represents the first decision variable between the i-th UAV and the j-th flight mission, x ij ∈{0,1},y ijk represents the second decision variable between the i-th UAV and the j-th flight mission and the k-th flight mission, y ijk ∈{0,1};C i represents the maximum load data of the i-th UAV, l j Represents the required material weight of the j-th flight mission.

[0070] Among them, the first decision variable specifically refers to the UAV U i Whether to execute the flight mission T j , if so, then the first decision variable x ij The value of is 1, otherwise, the first decision variable x ij The value of is 0. The second decision variable specifically refers to the drone U i Whether to start from the task point T i Fly to mission point T k , if so, then the second decision variable y ijk The value is 1, otherwise, the second decision variable y ijk The value is 0.

[0071] The adjustment coefficients λ1, λ2 and λ3 are used to achieve a balance between path length, task priority and load balancing. Based on the above constraints, the joint optimization model of this embodiment optimizes the path planning and task allocation of the drone cluster by adjusting the weight coefficients in the objective function.

[0072] It should be noted that the method of this embodiment also includes designing drone formation grouping, and the drone ID is uniquely identified by a decimal number to facilitate the management of the formation structure. At the same time, the drones are divided into multiple subgroups according to task requirements and locations, and each subgroup focuses on a specific task to improve execution efficiency.

[0073] Specifically, in complex application scenarios such as engineering construction, when drone clusters perform tasks, they need to be reasonably formed and grouped according to task requirements, locations, priorities, etc., in order to achieve efficient management and task coordination. By comparing decimal number representation and binary number representation, it is found that decimal numbers are easier to understand and retrieve quickly. Therefore, the present invention adopts decimal number representation as the unique identification scheme for drones, and at the same time divides the drone cluster into multiple subgroups in combination with task requirements, each subgroup focuses on a specific task to improve execution efficiency.

[0074] Assuming that the number of drones is 30, the present invention designs the ID number by three decimal numbers, such as Figure 2 As shown. The decimal number in the percentile position represents the number of the UAV formation, and the combination of the decimal numbers in the tenth and units positions represents the number of each UAV in the formation. Therefore, the format of the UAV ID designed using decimal numbers is UAV101, UAV214, etc., which represent UAV No. 1 in formation No. 1 and UAV No. 14 in formation No. 2, respectively.

[0075] Step three, construct a path node graph according to the grid map, update pheromones according to the Tent chaos map and ant colony algorithm, and combine the path optimization objective function to obtain the optimal path planning solution for the drone cluster.

[0076] The first step is to construct a path node graph based on the grid map.

[0077] Specifically, by analyzing the grid map, important feature points are extracted as nodes for path planning. The feature points are connected to form a searchable path graph, and a path node graph is constructed, such as Figure 4 As shown, the method for obtaining the path node graph can be implemented by steps S301 to S303.

[0078] Step S301, constructing a grid map based on the working environment of the drone cluster, wherein the element value of the obstacle in the grid map is 1, the element value of the task target is 2, and the element value of the free grid is 0.

[0079] In this embodiment, a grid map is used as the working environment of the drone cluster. In the grid map, black squares represent obstacles, and the element value in the corresponding matrix is ​​1, white squares represent free grids, and the element value in the corresponding matrix is ​​0, and gray squares represent mission targets, and the element value in the corresponding matrix is ​​2.

[0080] Step S302, obtaining a label of each grid in the grid map according to the grid number of each grid in the grid map and the horizontal coordinate scale and the vertical coordinate scale of the grid.

[0081] Specifically, the calculation formula of the label can be expressed as:

[0082]

[0083] Among them, x v and v They represent the labels corresponding to the grid number v, mod is the remaining operator, N x Indicates the horizontal scale of the grid, N y Represents the vertical scale of the grid, cell is the round-up function, and v is the grid number.

[0084] Step S303: In the grid map, a square composed of four adjacent grids is a basic unit. If only one grid in the basic unit is marked as an obstacle, the center point of the basic unit corresponding to the square is used as the key feature point, and the key feature point is used as the node of path planning to construct a path node graph.

[0085] It should be noted that in this embodiment, in order to avoid collisions during the flight of the drone cluster, the obstacles are appropriately expanded to the range of the three-layer grid to meet the safety distance requirements of the drone. The irregular obstacle area is standardized and projected into the grid map, and then the obstacle boundary is expanded according to the radius and safety distance of the drone, and the side length of the expanded area is equal to the flight radius of the drone. The expanded obstacle vertices are extracted as feature points, and some points do not need to be extracted, so the extraction condition of the feature points is set as follows: in a square composed of four adjacent grids, only one grid is marked as an obstacle, and its center point is the feature point; duplicate or redundant vertices are excluded, and only key obstacle points that affect path planning are retained.

[0086] Specifically, a square composed of four adjacent grids in the grid map is a basic unit, and then the expanded obstacle-related area can be judged. If only one grid in the basic unit corresponding to a square is marked as an obstacle, that is, when the element value corresponds to 1, the center point of the basic unit corresponding to the square is used as the key feature point. Through such conditional screening, points that can represent the key positions of obstacles and have an important impact on path planning can be found from many grids.

[0087] It should be noted that, in this embodiment, after extracting the feature points that initially meet the conditions, repeated or redundant vertices are further excluded, and only those key obstacle points that truly affect path planning, that is, the final key feature points, are retained.

[0088] Finally, these key feature points extracted are used as nodes for path planning. According to their positional relationship in the grid map and factors such as passability, for example, feature points corresponding to adjacent non-obstruction areas can be connected. These feature points are connected by the adjacency method combined with the feasible path method. Specifically, it is determined whether the feature points are directly adjacent. If there are no obstacles between the grids corresponding to the two points, a direct connection is established; for feature points that are not directly adjacent but close in distance, check whether the straight path between the two points passes through obstacles. If the path is completely passable, a connection is established, thereby constructing a path node graph. The path node graph presents the path network structure that can be selected for drone flight, and becomes the basic carrier for subsequent search for the optimal path.

[0089] The second step is to update pheromones based on Tent chaotic mapping and ant colony algorithm.

[0090] In this embodiment, the pheromone update mechanism is optimized, and an uneven pheromone distribution is adopted to make the better path have a higher pheromone concentration. Tent chaotic mapping is introduced in path selection to achieve randomness and diversity of path selection and avoid deadlock. The pheromone volatilization rate is dynamically adjusted to adapt to the complex and changeable flight environment.

[0091] It should be noted that, since the traditional ant colony algorithm is limited by the turning angle and has a high degree of randomness, the quality of the initial solution cannot be guaranteed. Therefore, the Tent chaotic mapping is used to optimize the ant colony algorithm, and the random chaotic sequence generated by it is used to optimize the initial ant colony. In addition, the traditional ant colony algorithm pheromone update mechanism, under the influence of the same pheromone concentration enhancement coefficient, the ants will not consider the priority order of tasks when searching for tasks. In the task context of this embodiment, this update mechanism may cause the drone formation to constantly switch formations when performing tasks, which increases the complexity of the problem. Therefore, if Figure 3 As shown, this embodiment combines the Tent chaotic mapping to optimize the ant colony algorithm.

[0092] In the initialization stage of path planning, the chaotic sequence X(t) generated by the Tent chaotic map is used to provide higher quality heuristic information for the initial position and distribution of the ants. The formula of the Tent map can be expressed as:

[0093]

[0094] Among them, X(t+1) represents the chaotic sequence value at time t+1, X(t) represents the chaotic sequence value at time t, X(t)∈(0,1), and t represents time t. The chaotic sequence value is used to guide the initial distribution of ants. The randomness of chaotic sequence generation ensures the diversity of ant path distribution in the initial stage and avoids too many ants from concentrating in the same area.

[0095] Then, based on the initial path optimized by the Tent chaotic mapping, the pheromone update process is optimized by taking the task priority factor into account in the pheromone update mechanism. Specifically, by assigning numbers to the drone formations and tasks, each formation corresponds to a task that matches its formation. When the ants are searching for tasks, if the number of the drone formation matches the task number, the pheromone concentration left by the ants will be higher; conversely, if the task number searched by the drone formation does not match, the pheromone concentration is lower, and the calculation formula of the pheromone enhancement coefficient can be expressed as:

[0096]

[0097] Among them, Q(i,j) represents the number of drones U i and flight mission T j The corresponding pheromone enhancement coefficient is as follows: i represents the UAV number, j represents the flight mission number, A represents the strengthening of pheromone update when the mission number and the UAV number match, and B represents the reducing of pheromone update when the mission number and the UAV number do not match.

[0098] Furthermore, when performing task assignment, the actual needs of path planning are taken into consideration. First, the task is randomly assigned to a node. Then, according to the current ant number, the critical point matrix takes the critical point corresponding to the ant number of different assignment strategies. Under the guidance of the pheromone enhancement factor calculation formula, the ants will spread pheromones and update the pheromone concentration. According to the pheromone rule, after each task is assigned to the drone formation, the path planning solution will be performed synchronously. During the task assignment process, once the drone obtains the task sequence, the path planning will generate a safe and flyable path for each drone and optimize the path by minimizing or maximizing the predefined objective function (such as seeking the shortest flight path).

[0099] Specifically, the pheromone update process can be expressed as:

[0100]

[0101] τ ij (t+n)=(1-ρ)·τ ij (t)+Δτ ij (t)

[0102]

[0103] Among them, τ ij (t) is the pheromone concentration between node i and node j at time t; ρ is the information volatility factor; α is the information heuristic factor; β is the expected heuristic factor; H(d ij ,Cost(U i ,T j)) is the comprehensive cost of UAV formation task allocation and path planning; Cost(U i ,T j ) is the cost of the ant colony selecting task i according to the difficulty of task execution, U i represents the i-th drone, T j represents the jth flight mission; d ij represents the distance between the i-th UAV and the j-th flight mission of the UAV, N m l represents the set of neighboring nodes of ant m at node l; Δτ ij m (t) represents the local increase of pheromone on edge (i, j) by ant m; Δτ ij (t) represents the total increase of pheromone of all ants on edge (i, j), p ij m represents the probability of ant m moving from node i to node j, τ ij (t+n) represents the pheromone concentration between node i and node j at time t+n.

[0104] The third step is to determine the optimal path planning scheme for the drone cluster based on the optimized ant colony algorithm and path optimization objective function, combined with the specific task requirements and locations of the drone subgroup. In addition, the drone path planning scheme of this embodiment can be dynamically adjusted to adapt to real-time environmental changes and achieve efficient task execution.

[0105] First, according to the optimized ant colony algorithm and path optimization objective function, the drone sub-groups are assigned tasks, and the optimal path planning scheme for the drone cluster is generated based on the specific task requirements and current position of each drone. Specifically, the system will reasonably assign tasks to different drones based on the priority of the task, the flight capability of the drone, the available resources, and the time constraints of the task. Each drone will calculate the best path according to the assigned task to ensure that the task can be completed in the most efficient way. The path planning process not only considers the spatial distribution of the task, but also evaluates the impact of environmental factors such as wind speed, obstacle location, and other dynamic factors on the flight path in real time. When generating the path, the goal is to optimize the path planning through predefined objective functions such as minimizing flight time, maximizing energy efficiency, or reducing flight distance. During the optimization process, the algorithm will ensure that the cluster maintains the best dynamic balance when performing tasks based on the overall synergy of the drone sub-group.

[0106] In addition, as the mission progresses and the environment changes, the path planning will be dynamically adjusted based on real-time data. For example, when the drone encounters an unexpected obstacle or needs to bypass certain areas, the system will immediately recalculate the path to ensure that the drone cluster can complete the mission efficiently and safely. This dynamic adjustment capability enables the drone cluster to flexibly respond to complex environmental changes and achieve high efficiency and reliability in mission execution.

[0107] In summary, the present invention combines intelligent algorithms to jointly optimize task allocation and path planning, taking into full account multiple factors such as flight obstacles, regional restrictions, and task priorities, and provides strong security and technical support for the efficient and safe flight of drone clusters in complex environments. This method not only reduces redundancy and conflict in the flight path, but also can dynamically adjust the flight path in real time to cope with environmental changes, ensuring the smooth completion of the task. By optimizing flight time and energy consumption, the present invention not only improves the execution efficiency of drone clusters, but also extends the endurance time and enhances its ability to adapt to complex mission scenarios, providing a solid technical foundation for the application of drones in monitoring, search and rescue, distribution and other fields.

[0108] The present invention also provides a drone cluster management system based on an intelligent algorithm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the computer program is executed by the processor, the steps of a drone cluster management method based on an intelligent algorithm are implemented. Since the implementation method of a drone cluster management method based on an intelligent algorithm has been described in detail above, it will not be repeated here.

[0109] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A drone cluster management method based on intelligent algorithm, characterized in that: The method comprises the following steps: Obtain the operating data of each drone in the drone cluster and the attribute data of each flight mission in various dimensions; According to the load distribution and time dimension attribute data in the operation data of each drone in the drone cluster, combined with the flight path of the drone, the constraint conditions and path optimization objective function are constructed; A path node graph is constructed according to the grid map, pheromones are updated according to the Tent chaotic map and ant colony algorithm, and the optimal path planning scheme of the UAV cluster is obtained by combining the path optimization objective function.

2. The method for managing drone clusters based on intelligent algorithms according to claim 1, characterized in that: The operation data of the UAV includes the maximum load data and the maximum cruising distance of the UAV; the attribute data of the flight mission in various dimensions includes the priority of the flight mission, the required material weight, the start time and end time of the flight mission, and the flight distance between different flight missions.

3. The method for managing drone clusters based on intelligent algorithms according to claim 2, characterized in that: The constraints specifically include: The weight of materials required for the corresponding flight mission performed by the drone shall not exceed the maximum load data of each drone; the mission time performed by the drone shall be between the start time and end time of the corresponding flight mission; the flight mission performed by the drone shall be continuous; the total flight distance of the drone shall not exceed the maximum endurance distance.

4. The method for managing drone clusters based on intelligent algorithms according to claim 3 is characterized in that: The path optimization objective function specifically includes: Wherein, F represents the path optimization objective function, λ1, λ2 and λ3 all represent preset adjustment coefficients, n represents the total number of drones, m represents the total number of drone flight missions, i represents the i-th drone, j represents the j-th flight mission, and k represents the k-th flight mission; d jk represents the flight distance between the jth flight mission and the kth flight mission of the UAV, p j represents the priority of the jth flight mission, x ij represents the first decision variable between the i-th UAV and the j-th flight mission, x ij ∈{0,1},y ijk represents the second decision variable between the i-th UAV and the j-th flight mission and the k-th flight mission, y ijk ∈{0,1};C i represents the maximum load data of the i-th UAV, l j Represents the required material weight of the j-th flight mission.

5. The method for managing drone clusters based on intelligent algorithms according to claim 1, characterized in that: The step of constructing a path node graph according to the grid map specifically includes: A grid map is constructed based on the working environment of the drone cluster, wherein the element value of the obstacle in the grid map is 1, the element value of the mission target is 2, and the element value of the free grid is 0; According to the grid number of each grid in the grid map and the horizontal coordinate scale and vertical coordinate scale of the grid, the label of each grid in the grid map is obtained; In the grid map, a square composed of four adjacent grids is the basic unit. If only one grid in the basic unit is marked as an obstacle, the center point of the basic unit corresponding to the square is used as the key feature point, and the key feature point is used as the node of path planning to construct a path node graph.

6. The method for managing drone clusters based on intelligent algorithms according to claim 5, characterized in that: The step of obtaining a label of each grid in the grid map according to the grid number of each grid in the grid map and the horizontal coordinate scale and the vertical coordinate scale of the grid specifically includes: Among them, x v and v They represent the labels corresponding to the grid number v, mod is the remaining operator, N x Indicates the horizontal scale of the grid, N y Represents the vertical scale of the grid, cell is the round-up function, and v is the grid number.

7. The method for managing drone clusters based on intelligent algorithms according to claim 4, characterized in that: The formula of the Tent chaotic mapping is specifically: Among them, X(t+1) represents the chaotic sequence value at time t+1, X(t) represents the chaotic sequence value at time t, X(t)∈(0,1), and t represents time t.

8. The method for managing drone clusters based on intelligent algorithms according to claim 7 is characterized in that: The updating of pheromones according to the Tent chaotic mapping and the ant colony algorithm specifically includes: t ij (t+n)=(1-ρ)·τ ij (t)+Δτ ij (t) Among them, τ ij (t) is the pheromone concentration between node i and node j at time t; ρ is the information volatility factor; α is the information heuristic factor; β is the expected heuristic factor; H(d ij ,Cost(U i ,T j )) is the comprehensive cost of UAV formation task allocation and path planning; Cost(U i ,T j ) is the cost of the ant colony selecting task i according to the difficulty of task execution, U i represents the i-th drone, T j represents the jth flight mission; d ij represents the distance between the i-th UAV and the j-th flight mission of the UAV, N m l represents the set of neighboring nodes of ant m at node l, Δτ ij m (t) represents the local increase of ant m’s pheromone on edge (i, j), Δτ ij (t) represents the total increase of pheromone of all ants on edge (i, j), p ij m represents the probability of ant m moving from node i to node j, τ ij (t+n) represents the pheromone concentration between node i and node j at time t+n.

9. A drone swarm management system based on intelligent algorithms, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the computer program is executed by the processor, the steps of the drone cluster management method based on intelligent algorithm as described in any one of claims 1 to 8 are implemented.

Images