Fixed-wing unmanned aerial vehicle group inspection task planning optimization method and system based on ICBBA-IGJO

Abstract

The invention belongs to the technical field of unmanned aerial vehicle group inspection task planning, and particularly relates to a fixed-wing unmanned aerial vehicle group inspection task planning optimization method and system based on ICBBA-IGJO, and the method comprises the steps: generating an optimal task distribution scheme for a fixed-wing unmanned aerial vehicle group through employing an improved consensus binding algorithm; and then, by taking the minimum total inspection cost of the fixed-wing unmanned aerial vehicle group as a target, constructing an unmanned aerial vehicle group inspection path optimization model, and based on the optimal task allocation scheme, solving the unmanned aerial vehicle group inspection path optimization model by utilizing an improved litsea rotundifolia optimization algorithm to obtain an optimal inspection path. According to the invention, joint optimization of task allocation and path planning of the fixed-wing unmanned aerial vehicle group is realized, so that the intelligent operation and maintenance level of the fixed-wing unmanned aerial vehicle group is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of inspection mission planning for unmanned aerial vehicle (UAV) swarms, and particularly relates to an optimization method, system, device and medium for inspection mission planning of fixed-wing UAV swarms based on ICBBA-IGJO. Background Art

[0002] Traditional inspection methods include manual inspection and helicopter inspection, etc. However, inspection tasks are often intricate, have a wide coverage area and are scattered. In addition, due to the complex terrain and inconvenient transportation in cities or mountainous areas, manual inspection is difficult and inefficient. Using helicopters is costly and not conducive to large-scale inspections. At the same time, as a modern aviation device, UAVs have played an important role in many fields. Their advantages such as low price, high mobility, strong timeliness, high-probability line-of-sight channels, and high positioning accuracy exactly match the inspection requirements. Currently, UAVs have been applied to a certain extent in power inspections, supporting the inspection strategy of "mainly using UAVs and supplemented by humans", and gradually developing towards fully unmanned inspections.

[0003] The literature "Three-Dimensional UAV Path Planning Based on Improved A* Algorithm" plans the three-dimensional path of UAVs based on the improved A* algorithm, improves the algorithm search efficiency by using variable-step search, and designs a path evaluation function with variable weights for optimization. However, the above method only solves the single-UAV path planning problem and the path search problem in two-dimensional scenarios. The literature "Method for Assigning Disaster Damage Inspection Tasks of Multiple UAVs in Distribution Networks Based on Improved Particle Swarm" assigns tasks based on biological intelligence methods. The above method belongs to a centralized setting. Although the logic is simple, it is not suitable for large-scale task assignment. The literature "Distributed multirobot task assignment via consensus ADMM" conducts distributed task assignment by introducing the alternating direction multiplier method, and proves through experiments that the proposed method can converge to the optimal solution in most cases. The above method can quickly provide an acceptable task assignment scheme, but does not consider the path planning problem. Therefore, there is an urgent need to develop an optimization method applicable to UAV swarm task assignment and path planning. Summary of the Invention

[0004] The purpose of the present invention is to provide an optimization method, system, device and medium for inspection mission planning of fixed-wing UAV swarms based on ICBBA-IGJO, which is applicable to UAV swarm task assignment and path planning, aiming at the above problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention provides an optimization method for inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO. The optimization method includes:

[0007] Using an improved consensus bundling algorithm to generate an optimal task allocation scheme for the fixed-wing UAV swarm;

[0008] Taking the minimum total inspection cost of the fixed-wing UAV swarm as the goal, constructing an optimization model for the inspection path of the UAV swarm;

[0009] Based on the optimal task allocation scheme, using an improved golden jackal optimization algorithm to solve the optimization model for the inspection path of the UAV swarm, and obtaining the optimal inspection path.

[0010] The improved consensus bundling algorithm includes:

[0011] S1. Task package construction stage: Constructing the data structures required by the algorithm for the UAVs in the fixed-wing UAV swarm. The data structures include the task package of the th UAV, execution path list, winner list, winner price list. The task package represents the task set assigned to the th UAV. The tasks inside are arranged in the order of task addition. The execution path list represents the task execution path of the th UAV. The tasks inside are arranged in the order of task execution. The winner list represents the set of UAV serial numbers that the th UAV believes have won the task. The winner price list represents the set of bid prices of the UAVs that the th UAV believes have won the task;

[0012]

[0012] S2. Task selection stage: Dividing the total task set into an effective task set and an ineffective task set . Taking the maximum total benefit of the fixed-wing UAV swarm as the goal, based on the greedy strategy, select tasks from the effective task set and add them to the task package ;

[0013] S3. Redundancy removal stage: Conducting validity tests on each task in the winner list . If the task fails the validity test, it means that the task If a consensus cannot be reached through the preset conflict resolution rules, the task is divided into the invalid task set ; if the task passes the validity test, it means that a consensus can be reached through the conflict resolution rules. Add the UAV serial number corresponding to this task in the winning price list to the valid winner list , and add the UAV bid price corresponding to this task in the winning price list to the valid winner price list , and then enter S4;

[0014] Perform the validity test according to the following formula:

[0015] ;

[0016] In the above formula, , are both index functions; is used to judge whether the winner or winning price of the nd UAV and the th UAV adjacent to it in the communication topology for the task can reach a local consensus through the conflict resolution rules; is used to judge whether the winner or winning price of all UAVs and the UAVs adjacent to them in the communication topology for the task can reach a local consensus through the conflict resolution rules;

[0017] S4. Conflict resolution stage: When the th UAV receives the communication data sent by the UAV adjacent to it in the communication topology, update its own timestamp and perform update, reset, replacement propagation or default operation based on the valid winner list , valid winner price , timestamp through the conflict resolution rules, and then update the data structure;

[0018] S5. Return to step S2 for iterative calculation until the total benefit of the fixed-wing UAV swarm reaches the maximum.

[0019] The total benefit of the fixed-wing UAV swarm is the sum of the benefits of all UAVs. The functional expression of the benefit of the UAV is:

[0020] ;

[0021] ;

[0022] In the above formula, For the benefit of the th drone; For the th drone to reach the task according to the execution path list required time; For the th drone to execute the tasks in the execution path list the income value; For the discount factor of the tasks in the execution path list the ; For the clustering benefit of the execution path list ; , are the clustering categories of points , point respectively; For the Euclidean distance between point and point ; For the Euclidean distance between point and the clustering center to which point belongs; Point represents the new task added from the task package to the execution path list, and point represents the last task in the task package or the drone slot; , are the gain constant and the attenuation constant respectively.

[0023] The total inspection cost of the fixed-wing UAV swarm includes the total flight length cost, average flight height cost, total time consumption cost, and total power consumption cost of the fixed-wing UAV swarm. The calculation formula for the total inspection cost of the fixed-wing UAV swarm is:

[0024] ;

[0025] ;

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] In the above formula, is the total inspection cost of the fixed-wing UAV swarm; is the total flight length cost of the fixed-wing UAV swarm; is the average flight altitude cost of the fixed-wing UAV swarm; is the total time consumption cost of the fixed-wing UAV swarm; is the total power consumption cost of the fixed-wing UAV swarm; and and and are the weight coefficients of the corresponding costs respectively; is the th UAV, , is the total number of UAVs in the fixed-wing UAV swarm; is the flight path from mission to mission ; is the Euclidean distance from mission to mission ; is the number of path points in path ; is a binary decision variable, which takes the value of 1 when both mission and mission belong to the mission sequence of UAV , otherwise it takes the value of 0; and respectively represent the Z-axis coordinates of the th and th path points in path ; is the distance conversion coefficient; is the total flight distance of UAV ; is the number of missions assigned to UAV after the mission assignment is completed; is the total flight time of UAV ; is the total number of missions of a single UAV; is the power consumption of UAV ; is the power per unit distance of the UAV;

[0032] The constraint conditions of the UAV swarm inspection path optimization model described above include:

[0033] Safety distance constraint:

[0034] ;

[0035] In the above formula, is the th UAV, , represents the distance between the UAV and the UAV or the distance between the UAV and the obstacle ; is the distance safety threshold.

[0036] The improved golden jackal optimization algorithm includes: first, determining the positions of the male and female jackals according to the fitness value, where the male jackal is the jackal with the highest fitness value in the golden jackal population, and the female jackal is the jackal with the second highest fitness value in the golden jackal population; then, determining the position update strategy according to the prey escape energy If the prey escape energy satisfies , then the exploration strategy is adopted to update the relative positions of all jackal individuals and the prey, otherwise the exploitation strategy is adopted to update the relative positions of all jackal individuals and the prey;

[0037] The formula for calculating the prey escape energy is:

[0038] ;

[0039] ; ;

[0040] In the above formula, , are the escape random factor and the iteration number decay factor respectively; is a random value in (0, 1); is a constant; is the current iteration number; is the maximum iteration number.

[0041] The formula for updating the relative position in the exploration strategy is:

[0042] ;

[0043] ;

[0044] ;

[0045] ;

[0046] In the above formula, , represent the relative position vectors of the male and female jackals with respect to the prey at the th iteration respectively; , represent the relative position vectors of the male and female jackals with respect to the prey at the The position vector at the th iteration; The position vector of the prey at the th iteration; Indicates the random number vector based on the Lévy distribution; For the relative position vector between the golden jackal individual and the prey at the th iteration; Is the proportional balance factor; Is a constant; Is a random value in (0, 1).

[0047] The relative position update formula of the exploitation strategy is improved by using the golden jackal observation strategy. The improved relative position update formula of the exploitation strategy is:

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] In the above formula, , respectively represent the relative position vectors between the male jackal and the female jackal and the prey at the th iteration; , respectively represent the position vectors of the male jackal and the female jackal at the th iteration; , respectively represent the position vectors of the male jackal and the female jackal after observation; , are the upper and lower bounds of the position respectively; Is the attenuation factor; Is the maximum number of iterations; Is a random number following a normal distribution; Is a random number uniformly distributed between 0 and 1; For the relative position vector between the golden jackal individual and the prey at the th iteration; Is the proportional balance factor; Is a constant; is a random value in (0, 1).

[0056] In a second aspect, the present invention provides a fixed-wing UAV group inspection task planning optimization system based on ICBBA-IGJO, wherein the optimization system includes a task allocation module, a model building module, and a path optimization module;

[0057] The task allocation module is used to generate an optimal task allocation plan for the fixed-wing UAV group using an improved consensus bundling algorithm;

[0058] The model building module is used to build a drone group inspection path optimization model with the goal of minimizing the total inspection cost of the fixed-wing drone group;

[0059] The path optimization module is used to solve the drone group inspection path optimization model based on the optimal task allocation plan and use the improved golden jackal optimization algorithm to obtain the optimal inspection path.

[0060] The task allocation module includes a task package construction submodule, a task selection submodule, a redundancy removal submodule, and a conflict resolution submodule;

[0061] The task package construction submodule is used to construct the data structure required by the algorithm for the drones in the fixed-wing drone group, and the data structure includes the first Mission package for drones , Execution path list , Winners List , Winning Price List , the task package Indicates that the A collection of drone missions. The tasks are arranged in the order in which they are added, and the execution path list Indicates The mission execution path of the UAV, The tasks are arranged in the order of task execution, and the winner list Indicates storage The drone serial number set that the drone believes won the mission, the winning price list Indicates The set of bid prices of the drones that the drones believe will win the mission;

[0062] The task selection submodule is used to set the total task Divide into effective task sets and invalid task collection , with the goal of maximizing the total benefit of the fixed-wing UAV swarm, based on the greedy strategy, from the effective task set Select the task to add to the task package middle;

[0063] The redundancy removal sub-module is used to perform validity checks on each task in the winner list. If a task fails the validity check, it means that the task cannot reach a consensus through the preset conflict resolution rules, and the task is classified into the invalid task set ; if a task passes the validity check, it means that it can reach a consensus through the conflict resolution rules. The UAV serial number corresponding to this task in the winning price list is added to the valid winner list , and the UAV bid price corresponding to this task in the winning price list is added to the valid winner price list ; the formula for the validity check is:

[0064] ;

[0065] In the above formula, , are both indicator functions; is used to determine whether the winner or winning price of the th UAV and the th UAV adjacent to it in the communication topology for the task can reach a local consensus through the conflict resolution rules; is used to determine whether the winner or winning price of all UAVs and the UAVs adjacent to them in the communication topology for the task can reach a local consensus through the conflict resolution rules;

[0066] The conflict resolution sub-module is used to update its own timestamp when the th UAV receives communication data sent by the UAVs adjacent to it in the communication topology, and based on the valid winner list , the valid winner price , and the timestamp , perform update, reset, replacement propagation, or default operations through the conflict resolution rules. Subsequently, the data structure is updated, and the updated data structure is input into the task selection sub-module for continuous iterative calculation until the total benefit of the fixed-wing UAV swarm reaches the maximum. The total benefit of the fixed-wing UAV swarm is the sum of the benefits of all UAVs. The functional expression of the benefit of the UAV is:

[0067]

[0068] ;

[0069] ;

[0070] In the above formula, For the The benefits of flying drones; For the The drones follow the execution path list Arrival Mission the time required; For the List of execution paths for drones Medium Task The revenue value; List of execution paths Medium Task Discount factor; List of execution paths The clustering benefits of , Points ,point The clustering category of For point With point The Euclidean distance between For point With point The Euclidean distance between the cluster centers to which they belong; points Indicates a new task added to the execution path list from the task package. Click Indicates the last mission or drone slot in a mission package; , are the gain constant and the attenuation constant respectively.

[0071] The total inspection cost of the fixed-wing UAV group includes the total flight length cost, average flight height cost, total time consumption cost, and total power consumption cost of the fixed-wing UAV group. The calculation formula of the total inspection cost of the fixed-wing UAV group is:

[0072] ;

[0073] ;

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] In the above formula, is the total inspection cost of the fixed-wing UAV swarm; is the total flight length cost of the fixed-wing UAV swarm; is the average flight altitude cost of the fixed-wing UAV swarm; is the total time consumption cost of the fixed-wing UAV swarm; is the total power consumption cost of the fixed-wing UAV swarm; , , , are the weight coefficients of the corresponding costs respectively; is the th UAV, , is the total number of UAVs in the fixed-wing UAV swarm; is from task to task 's flight path; is from task to task 's Euclidean distance; is the number of path points in path ; is a binary decision variable, which takes the value of 1 when both task and task belong to the task sequence of UAV , otherwise it takes the value of 0; , respectively represent the Z-axis coordinates of the th and th path points in path ; is the distance conversion coefficient; is the total flight distance of UAV ; is the number of tasks assigned to UAV after task allocation; is the total flight time of UAV ; is the total number of tasks of a single UAV; is the power consumption of UAV ; is the power per unit distance of the UAV;

[0080] The constraint conditions of the UAV swarm inspection path optimization model include:

[0081] Safety distance constraint:

[0082] ;

[0083] In the above formula, is the th unmanned aerial vehicle, , represents the distance between the unmanned aerial vehicle and the unmanned aerial vehicle , or the distance between the unmanned aerial vehicle and the obstacle ; is the distance safety threshold.

[0084] The improved golden jackal optimization algorithm includes: first, determining the positions of the male jackal and the female jackal according to the fitness value, where the male jackal is the jackal with the highest fitness value in the golden jackal population, and the female jackal is the jackal with the second highest fitness value in the golden jackal population; then, determining the position update strategy according to the prey escape energy . If the prey escape energy satisfies , then the exploration strategy is adopted to update the relative positions of all jackal individuals and the prey, otherwise the exploitation strategy is adopted to update the relative positions of all jackal individuals and the prey;

[0085] The calculation formula of the prey escape energy is:

[0086] ;

[0087] ; ;

[0088] In the above formula, , are the escape random factor and the iteration number attenuation factor respectively; is a random value in (0, 1); is a constant; is the current iteration number; is the maximum iteration number.

[0089] The relative position update formula in the exploration strategy is:

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] In the above formula, , respectively represent the relative position vectors of the male jackal and the female jackal with respect to the prey at the th iteration; , respectively represent the position vectors of the male and female jackals at the -th iteration; represents the position vector of the prey at the -th iteration; represents the escape coefficient of the prey; represents a random number vector based on the Lévy distribution; is the relative position vector between the golden jackal individual and the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1).

[0095] The path optimization module is also used to improve the relative position update formula of the exploitation strategy by using the golden jackal observation strategy. The improved relative position update formula of the exploitation strategy is:

[0096] ;

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] ;

[0103] In the above formula, , respectively represent the relative position vectors between the male and female jackals and the prey at the -th iteration; , respectively represent the position vectors of the male and female jackals at the -th iteration; , respectively represent the position vectors of the male and female jackals after observation; , are the upper and lower bounds of the position respectively; is the attenuation factor; is the maximum number of iterations; is a random number following a normal distribution; is a random number uniformly distributed between 0 and 1; is the relative position vector between the golden jackal individual and the prey at the The relative position vector with the prey at the i-th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1).

[0104] In a third aspect, the present invention also provides an optimization device for inspection task planning of a fixed-wing UAV swarm based on ICBBA-IGJO. The optimization device includes a memory and a processor; the memory is used to store computer program code and transmit the computer program code to the processor; the processor is used to execute the foregoing optimization method according to the instructions in the computer program code.

[0105] In a fourth aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the foregoing optimization method is implemented.

[0106] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0107] 1. For the optimization method for inspection task planning of a fixed-wing UAV swarm based on ICBBA-IGJO of the present invention, first use the improved consensus bundling algorithm to generate an optimal task allocation plan for the fixed-wing UAV swarm, and then, with the goal of minimizing the total inspection cost of the fixed-wing UAV swarm, construct an optimization model for the inspection path of the UAV swarm. Based on the optimal task allocation plan, use the improved golden jackal optimization algorithm to solve the optimization model for the inspection path of the UAV swarm to obtain the optimal inspection path; the above method realizes the joint optimization of task allocation and path planning of the fixed-wing UAV swarm, thereby improving the intelligent operation and maintenance level of the fixed-wing UAV swarm. Therefore, the present invention can realize the joint optimization of task allocation and path planning of the fixed-wing UAV swarm and improve the intelligent operation and maintenance level of the fixed-wing UAV swarm.

[0108] 2. For the optimization method for inspection task planning of a fixed-wing UAV swarm based on ICBBA-IGJO of the present invention, a redundancy removal stage is added to the task package construction stage in the traditional consensus bundling algorithm, and each task in the winner list is subjected to an effectiveness test. The tasks that fail the effectiveness test are divided from the effective task set to the invalid task set, so as to avoid continuing to process invalid tasks in the task selection stage in subsequent iterations, save computing and communication bandwidth resources, reduce the consensus cost, and accelerate the algorithm convergence speed. Therefore, the present invention reduces the consensus cost and accelerates the algorithm convergence speed through the redundancy removal stage in task allocation optimization.

[0109] 3. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to the present invention uses a category-based reward system to improve the function expression of the UAV benefit. In the category-based reward system, the reward is designed according to whether the new task added to the execution path list from the task package and the last task point or UAV slot in the task path list belong to the same cluster, which can prevent the UAV from adding tasks at a relatively long distance due to the greedy strategy, resulting in a reduction in the total benefit of the fixed-wing UAV swarm and an increase in the total path length and total task time. Therefore, the present invention can ensure the total benefit of the fixed-wing UAV swarm in task assignment optimization and reduce the total path length and total task time.

[0110] 4. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to the present invention, in the improved golden jackal optimization algorithm, first determines the position update strategy according to the prey escape energy, which can adaptively adjust the search range, avoid falling into local optima, and improve the search accuracy; secondly, a proportional balance factor is introduced into the position update formula to randomly adjust the contribution ratio of the male jackal and the female jackal in each update process, providing more jumping possibilities and preventing all individuals in the population from tending to the same solution, thereby reducing the risk of falling into local optima; furthermore, in the relative position update formula using the strategy, a dynamic migration mechanism and a dynamic feedback migration mechanism are introduced to further enhance the ability of the population to jump out of the local optimum value. Therefore, the present invention can avoid falling into local optima during path planning optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 is the flowchart of the optimization method described in the present invention.

[0112] Figure 2 is the path optimization result of the optimization method proposed in the present invention.

[0113] Figure 3 is the structural block diagram of the optimization system described in the present invention.

[0114] Figure 4 is the structural block diagram of the optimization device described in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0115] The present invention will be further described in detail below in conjunction with the specific embodiments and the accompanying drawings.

[0116] Example 1:

[0117] Refer to Figure 1 , an optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO, which is carried out in the following steps in sequence:

[0118] Step 1: Use the Improved Consensus Bundling Algorithm (ICBBA) to generate an optimal task allocation scheme for a fixed-wing UAV swarm; The traditional Consensus Bundling Algorithm (CBBA) is divided into two stages. The first stage is the task package construction stage, and the second stage is the conflict resolution stage; For the improved consensus bundling algorithm of the present invention, first, a redundancy removal stage is added after the first stage to exclude potential conflict-free tasks in advance; Secondly, the UAV benefit function is improved using a category-based reward system; Finally, a new conflict resolution rule is established in the conflict resolution stage to achieve task allocation; The improved consensus bundling algorithm includes:

[0119] S1. Task package construction stage: Construct the data structures required by the algorithm for the UAVs in the fixed-wing UAV swarm. The data structures include the task package of the th UAV, the execution path list the winner list and the winner price list . The task package represents the set of tasks assigned to the th UAV. The tasks inside are arranged in the order of task addition. The execution path list represents the task execution path of the th UAV. The tasks inside are arranged in the order of task execution. The winner list represents the set of UAV serial numbers that the th UAV believes have won the tasks. Its size is a matrix of . is the number of tasks currently being allocated. The winner price list represents the set of UAV bid prices that the th UAV believes have won the tasks. Each of its elements corresponds to ; Example: Suppose there are currently 5 tasks being allocated {T1, T2, T3, T4, T5}, and the winner list is [2, 1, 3, 5, 4]. Then is 5, and the UAV serial numbers of the winners of each task are 2, 1, 3, 5, and 4 respectively;

[0120] S2. Task selection stage: Divide the total task set into a valid task set and an invalid task set . The UAV selects tasks from the valid task set and adds them to the task package In order to maximize the total benefit of the fixed-wing UAV swarm, where the total benefit of the fixed-wing UAV swarm is the sum of the benefits of all UAVs; to prevent UAVs from adding tasks at relatively long distances due to the greedy strategy, which reduces the total benefit of the fixed-wing UAV swarm and increases the total path length and total task time, the function expression of the UAV benefit is improved through a category-based reward system. In the category-based reward system, the task points and the UAV nest are grouped without distinction using the Kmeans algorithm. The clustering results are used to determine whether the new task added from the task package to the execution path list and the last task point in the task path list belong to the same cluster category. If the task path list is empty, it is determined whether the new task added from the task package to the execution path list and the UAV nest belong to the same category; the function expression of the UAV benefit is obtained as follows:

[0121] ;

[0122] ;

[0123] In the above formula, is the benefit of the th UAV; is the time required for the th UAV to reach task according to the execution path list ; is the revenue value of task executed by the th UAV in the execution path list ; is the discount factor of task in the execution path list ; is the clustering benefit of the execution path list ; , are the clustering categories of point and point respectively. indicates that the clustering category of point is the same as that of point . Conversely, indicates that the clustering category of point is different from that of point ; is the Euclidean distance between point and point ; is the Euclidean distance between point and the clustering center to which point belongs; point represents the new task added from the task package to the execution path list, and point Indicates the last task in the task package or the drone slot; and are the gain constant and the attenuation constant respectively;

[0124] S3, redundancy removal phase: For each task in the winner list perform validity checks according to the following formula:

[0125] ;

[0126] In the above formula, and are both index functions; is used to determine whether the th drone and the th drone adjacent to it in the communication topology can reach a local consensus on the winner or winning price for the task through the conflict resolution rule; is used to determine whether all drones and the drones adjacent to them in the communication topology can reach a local consensus on the winner or winning price for the task through the conflict resolution rule;

[0127] If the task fails the validity check, it means that the task cannot reach a consensus through the preset conflict resolution rule. The task is divided from the valid task set to the invalid task set , and added to the invalid task set At the end, it avoids the task selection phase in subsequent iterations from continuing to process such invalid tasks, saving computing and communication bandwidth resources, avoiding unnecessary information transmission, and accelerating the algorithm convergence speed. Invalid tasks include but are not limited to the following situations: Situation 1: The resources required by the task, such as time and priority, cannot be satisfied under the current conditions; Situation 2: The execution order or dependency relationship between tasks is not satisfied; Situation 3: The flight ability of the UAV cannot meet the task. To ensure that all tasks can be allocated, invalid tasks may still become valid tasks during subsequent iterations, be included in the valid task set again, and participate in the allocation again. The situations where invalid tasks may become valid tasks include but are not limited to: 1. When the status or conditions of the task change, it may become valid again. For example, the priority between tasks is satisfied in this validity check; 2. A certain task was invalid before due to reasons such as insufficient UAV flight ability or task constraint conflicts. Now, due to the completion, cancellation of other tasks or the change of the UAV status, resources are released. These resources may make the previously invalid task become valid again; 3. The execution conditions of a certain task become more suitable for the current UAV's ability. For example, the time window of the task is adjusted or the distance between the task and the UAV becomes suitable for allocation. At this time, the task can be changed from invalid to valid;

[0128] If the task passes the validity check, it means that consensus can be reached through the conflict resolution rule and it needs to continue to participate in subsequent communication exchanges. At this time, the UAV serial number corresponding to this task in the winning price list is added to the valid winner list and the UAV bid price corresponding to this task in the winning price list is added to the valid winner price list , and then it enters S4;

[0129] S4. After removing redundant tasks, each UAV will use the valid winner list , valid winner price , time stamp to spread task information through the conflict resolution rule. The conflict resolution phase is specifically as follows:

[0130] First, when the th UAV receives communication data sent by a UAV adjacent to it in the communication topology, it updates its own time stamp :

[0131] ;

[0132] In the above formula, is the information reception time; represents the The element value in the th row and th column represents the communication quality between the th drone (recipient) and the th drone (sender). The larger the value, the more communication loss; is the minimum value in the th row of the communication matrix. Let this minimum value point to the th drone; represents the timestamp pointed to by the maximum value in the th row of the communication matrix. Let this maximum value point to the th drone. Then represents the timestamp of the th drone when it receives the communication data sent by the

[0133] Then, based on the valid winner list , the valid winner price , and the timestamp , update, reset, replace and propagate or default retain through the conflict resolution rules to propagate the task information, and then update the data structure;

[0134] The conflict resolution rules are as described in Table 1. In Table 1:

[0135] Update: ; ; , are the valid winners of the task considered by the th drone and the th drone respectively; , are the valid winner prices of the task considered by the th drone and the th drone respectively;

[0136] Reset: ; ;

[0137] Replace and propagate: The th drone is the sender;

[0138] Default retain: ; And wait for other drone senders to send information;

[0139] Table 1 Conflict Resolution Rules Table

[0140]

[0141] Where , are the timestamps of the UAVs when receiving communication data from the UAVs respectively; UAV ;

[0142] S5. Return to step S2 for iterative calculation until the total benefit of the fixed-wing UAV swarm reaches the maximum;

[0143] Step 2: Aiming at minimizing the total inspection cost of the fixed-wing UAV swarm, an optimization model for the inspection path of the UAV swarm is constructed. When the fixed-wing UAV swarm performs the inspection task, the allocation of its collaborative tasks is mainly affected by five factors, expressed as ; among them represents the inspection task environment of the distribution network. The UAV swarm performs tasks in the environment , being the three-dimensional coordinates; is a given fixed-wing UAV swarm, in indicating the th UAV; represents the UAV task set. Each UAV has a task set, and the task set of the th UAV can be expressed as , representing the total number of tasks of a single UAV; represents the type of tasks that the fixed-wing UAV swarm needs to perform; represents the UAV power, and the power of the th UAV is expressed as ;

[0144] The optimization model for the inspection path of the UAV swarm includes an objective function and a safety distance constraint. The objective function is to minimize the total inspection cost of the fixed-wing UAV swarm. The total inspection cost of the fixed-wing UAV swarm includes the total flight length cost, average flight height cost, total time consumption cost, and total power consumption cost of the fixed-wing UAV swarm. The calculation formula for the total inspection cost of the fixed-wing UAV swarm is:

[0145] ;

[0146] In the above formula, is the total inspection cost of the fixed-wing UAV swarm; is the total flight length cost of the fixed-wing UAV swarm; is the average flight height cost of the fixed-wing UAV swarm; is the total time consumption cost of the fixed-wing UAV swarm; is the total power consumption cost of the fixed-wing UAV swarm; , , , are the weight coefficients corresponding to the respective costs;

[0147] Since during the inspection process of a fixed-wing UAV swarm, a shorter flight path can save more energy and time costs, and a stable flight altitude will reduce the probability of multiple UAVs colliding with obstacles; , can be expressed as:

[0148] ;

[0149] ;

[0150] In the above formula, is the th UAV, , is the total number of UAVs in the fixed-wing UAV swarm; is the flight path from mission to mission ; is the Euclidean distance from mission to mission ; is the number of path points (i.e., missions) in path ; is a binary decision variable that takes the value of 1 when both mission and mission belong to the mission sequence of UAV , and takes the value of 0 otherwise; , respectively represent the Z-axis coordinates of the th and th path points in path ;

[0151] Since each mission has a time limit, it is necessary to limit the inspection time of the UAVs; can be expressed as:

[0152] ;

[0153] ;

[0154] In the above formula, is the distance conversion coefficient; is the total flight length of UAV ; is the number of missions assigned to UAV after the mission assignment is completed; is the UAV The total flight time; is the total number of tasks for a single UAV;

[0155] Considering the power limit of the UAV, it is necessary to constrain the power consumption of the UAV; It can be expressed as:

[0156] ;

[0157] ;

[0158] In the above formula, is the power consumption of the UAV ; is the power per unit distance of the UAV;

[0159] The safety distance constraint is:

[0160] ;

[0161] In the above formula, is the th UAV, ; represents the distance between the UAV and the UAV or the distance between the UAV and the obstacle ; is the distance safety threshold;

[0162] Step 3: Based on the optimal task allocation scheme, use the improved golden jackal optimization algorithm (IGJO) to solve the UAV swarm inspection path optimization model and obtain the optimal inspection path;

[0163] The improved golden jackal optimization algorithm includes: first, determine the positions of the male and female jackals according to the fitness value. The male jackal is the jackal with the highest fitness value in the golden jackal population, and the female jackal is the jackal with the second highest fitness value in the golden jackal population; then, determine the position update strategy according to the prey escape energy , adaptively adjust the search range, avoid falling into local optima and improve the search accuracy; if the prey escape energy satisfies , then adopt the exploration strategy to update the relative positions of all jackal individuals and the prey. The exploration strategy tends to make larger jumps, making the jackal individuals tend to search in farther areas, thus avoiding the population falling into local extrema; if the prey escape energy , then adopt the exploitation strategy to update the relative positions of all jackal individuals and the prey. The exploitation strategy makes the jackal individuals conduct more refined searches near the current position, thus improving the accuracy of the local solution; the calculation formula of the prey escape energy is:

[0164] ;

[0165] ; ;

[0166] In the above formula, and are the escape random factor and the iteration number attenuation factor respectively; is a random value in (0, 1); is a constant; is the current iteration number; is the maximum iteration number;

[0167] The relative position update formula in the exploration strategy is:

[0168] ;

[0169] ;

[0170] ;

[0171] ;

[0172] In the above formula, and represent the relative position vectors of the male jackal and the female jackal with respect to the prey at the -th iteration respectively; and represent the position vectors of the male jackal and the female jackal at the -th iteration respectively; represents the position vector of the prey at the -th iteration; represents the escape coefficient of the prey; represents a random number vector based on the Lévy distribution; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1);

[0173] The relative position update formula of the exploitation strategy is improved by using the golden jackal observation strategy. The improved relative position update formula of the exploitation strategy is:

[0174] ;

[0175] ;

[0176] ;

[0177] ;

[0178] ;

[0179] ;

[0180] ;

[0181] In the above formula, and respectively represent the relative position vectors of the male and female dholes with respect to the prey at the -th iteration; and respectively represent the position vectors of the male and female dholes at the -th iteration; and respectively represent the position vectors of the male and female dholes after observation; and are the upper and lower bounds of the position respectively; is the attenuation factor; is the maximum number of iterations; is a random number following a normal distribution; is a random number uniformly distributed between 0 and 1; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1);

[0182] In the relative position update formulas of the exploration strategy and the exploitation strategy, by introducing the proportional balance factor , the contribution ratio of and in each update process is random, providing more jumping possibilities. Even in the optimization within a small range, global exploration can be carried out through the guidance of sub-optimal solutions, increasing the diversity of solutions and avoiding the situation where all individuals in the population tend to the same solution, thereby reducing the risk of falling into local optima;

[0183] In the relative position update formula of the exploitation strategy, and They respectively represent the dynamic migration mechanism and the dynamic feedback migration mechanism; the dynamic migration mechanism is used to guide the individual golden jackals to transfer from the local optimal area to other broader areas, so as to explore more solution spaces and prevent the individual golden jackals from staying near the local optimal solution for a long time; the dynamic feedback migration mechanism is to adjust the search strategy by feedback the current search effect during the optimization process, making it easier for the individual golden jackals to leave the current local optimal solution at the right time and search in the direction of the global optimal solution; by introducing these two mechanisms, the ability of the population to jump out of the local optimal value is enhanced.

[0184] Performance verification:

[0185] Compare the optimization method proposed in this invention with the particle swarm algorithm, the mixed integer linear programming method, the ant colony algorithm, and reinforcement learning, and calculate the total inspection cost of the fixed-wing UAV swarm. The results are shown in Table 2:

[0186] Table 2 Comparison of the total inspection cost of the fixed-wing UAV swarm

[0187]

[0188] It can be seen from Table 2 that the total inspection cost of the fixed-wing UAV swarm of the optimization method proposed in this invention is the lowest. Compared with the particle swarm algorithm, the mixed integer linear programming method, the ant colony algorithm, and reinforcement learning, it leads by 8454, 6005, 5992, and 4001 respectively. The results show that the optimization method proposed in this invention can greatly shorten the total inspection cost of the UAV. The path optimization result of the optimization method proposed in this invention is as Figure 2 shown. It can be seen from Figure 2 that the flight length of each UAV first increases and then shortens during the iteration process and completes the iteration at about two hundred times. The results show that the optimization method proposed in this invention can effectively shorten the inspection path length of the UAV.

[0189] Example 2:

[0190] See Figure 3 , a fixed-wing UAV swarm inspection task planning optimization system based on ICBBA-IGJO, including a task allocation module, a model construction module, and a path optimization module; the task allocation module is used to generate an optimal task allocation plan for the fixed-wing UAV swarm by using the improved consensus bundling algorithm, the model construction module is used to construct an inspection path optimization model of the UAV swarm with the goal of minimizing the total inspection cost of the fixed-wing UAV swarm, and the path optimization module is used to solve the inspection path optimization model of the UAV swarm based on the optimal task allocation plan by using the improved golden jackal optimization algorithm to obtain the optimal inspection path;

[0191] Specifically, the task allocation module includes a task package construction sub-module, a task selection sub-module, a redundancy removal sub-module, and a conflict resolution sub-module. The task package construction sub-module is used to construct the data structures required by the algorithms for the unmanned aerial vehicles (UAVs) in the fixed-wing UAV swarm. The data structures include the task package of the th UAV , the execution path list , the winner list , the winning price list . The task package represents the set of tasks assigned to the th UAV. The tasks inside are arranged in the order of task addition. The execution path list represents the task execution path of the th UAV. The tasks inside are arranged in the order of task execution. The winner list represents the set of UAV serial numbers that the th UAV believes have won the task. The winning price list represents the set of bidding prices of the UAVs that the th UAV believes have won the task;

[0192] The task selection sub-module is used to divide the total task set into a valid task set and an invalid task set . With the goal of maximizing the total benefit of the fixed-wing UAV swarm, tasks are selected from the valid task set and added to the task package based on the greedy strategy. The total benefit of the fixed-wing UAV swarm is the sum of the benefits of all UAVs. The functional expression of the benefit of the UAV is:

[0193] ;

[0194] ;

[0195] In the above formula, is the benefit of the th UAV; is the time required for the th UAV to reach the task according to the execution path list ; is the benefit value of the th UAV for executing the task in the execution path list ; is the task in the execution path list ​ Discount factor; Is the execution path list Clustering gain; 、 Are the points 、 Clustering categories; Is the point And the point Euclidean distance between; Is the point And the point Euclidean distance between the cluster centers to which they belong; Point Represents a new task added from the task package to the execution path list, and the point Represents the last task in the task package or the drone slot; 、 Are the gain constant and the decay constant respectively;

[0196] The redundancy removal sub-module is used to perform validity checks on each task in the winner list . If the task Does not pass the validity check, it means that the task Cannot reach a consensus through the preset conflict resolution rules, and the task Is divided into the invalid task set ; If the task Passes the validity check, it means that a consensus can be reached through the conflict resolution rules, and the drone serial number corresponding to this task in the winning price list Is added to the valid winner list , and the drone bid price corresponding to this task in the winning price list Is added to the valid winner price list ; The formula for the validity check is:

[0197] ;

[0198] In the above formula, 、 Are both index functions; Is used to judge whether the st drone and the rd drone adjacent to it in the communication topology can reach a local consensus on the winner or winning price for the task ; Is used to judge whether all drones and the drones adjacent to them in the communication topology can reach a local consensus on the winner or winning price for the task ;

[0199] The conflict resolution sub-module is used to update its own timestamp when the nth fixed-wing UAV receives communication data sent by an adjacent UAV in the communication topology, and based on the valid winner list, valid winner price, timestamp, perform update, reset, replacement propagation or default operation through conflict resolution rules, then update the data structure, and input the updated data structure into the task selection sub-module for continuous iterative calculation until the total benefit of the fixed-wing UAV swarm reaches the maximum;

[0200] Specifically, the UAV swarm inspection path optimization model constructed by the model construction module includes an objective function and constraint conditions. The objective function is to minimize the total inspection cost of the fixed-wing UAV swarm. The total inspection cost of the fixed-wing UAV swarm includes the total flight length cost, average flight height cost, total time consumption cost, and total power consumption cost of the fixed-wing UAV swarm. The calculation formula for the total inspection cost of the fixed-wing UAV swarm is:

[0201] ;

[0202] ;

[0203] ;

[0204] ;

[0205] ;

[0206] ;

[0207] ;

[0208] In the above formula, is the total inspection cost of the fixed-wing UAV swarm; is the total flight length cost of the fixed-wing UAV swarm; is the average flight height cost of the fixed-wing UAV swarm; is the total time consumption cost of the fixed-wing UAV swarm; is the total power consumption cost of the fixed-wing UAV swarm; , , , are the weight coefficients of the corresponding costs respectively; is the th UAV, , is the total number of UAVs in the fixed-wing UAV swarm; For the flight path from task to task ; For the Euclidean distance from task to task ; Is the number of path points in path ; Is a binary decision variable, which takes the value of 1 when both task and task belong to the task sequence of the UAV , otherwise it takes the value of 0; , respectively represent the Z-axis coordinates of the th and th path points in path ; Is the distance conversion coefficient; Is the total flight distance of the UAV ; Is the number of tasks assigned to the UAV after the task assignment is completed; Is the total flight time of the UAV ; Is the total number of tasks of a single UAV; Is the power consumption of the UAV ; Is the power per unit distance of the UAV;

[0209] The above-mentioned constraint conditions include a safety distance constraint:

[0210] ;

[0211] In the above formula, Is the th UAV, , represents the distance between the UAV and the UAV or the distance between the UAV and the obstacle ; Is the distance safety threshold;

[0212] Specifically, the improved golden jackal optimization algorithm includes: first, determine the positions of the male jackal and the female jackal according to the fitness value. The male jackal is the jackal with the highest fitness value in the golden jackal population, and the female jackal is the jackal with the second highest fitness value in the golden jackal population; then, determine the position update strategy according to the prey escape energy . If the prey escape energy satisfies , the exploration strategy is adopted to update the relative positions of all jackal individuals and the prey, otherwise the exploitation strategy is adopted to update the relative positions of all jackal individuals and the prey;

[0213] The escape energy of the prey The calculation formula is:

[0214] ;

[0215] ; ;

[0216] In the above formula, , are the escape random factor and the iteration number decay factor respectively; is a random value in (0, 1); is a constant; is the current iteration number; is the maximum iteration number;

[0217] The relative position update formula in the exploration strategy is:

[0218] ;

[0219] ;

[0220] ;

[0221] ;

[0222] In the above formula, , respectively represent the relative position vectors of the male jackal and the female jackal with respect to the prey at the -th iteration; , respectively represent the position vectors of the male jackal and the female jackal at the -th iteration; represents the position vector of the prey at the -th iteration; represents the escape coefficient of the prey; represents a random number vector based on the Lévy distribution; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1);

[0223] Specifically, the path optimization module is further configured to improve the relative position update formula of the exploitation strategy by using the golden jackal observation strategy. The improved relative position update formula of the exploitation strategy is as follows:

[0224] ;

[0225] ;

[0226] ;

[0227] ;

[0228] ;

[0229] ;

[0230] ;

[0231] In the above formula, and respectively represent the relative position vectors of the male jackal and the female jackal with respect to the prey at the -th iteration; and respectively represent the position vectors of the male jackal and the female jackal at the -th iteration; and respectively represent the position vectors of the male jackal and the female jackal after observation; and are the upper and lower bounds of the position, respectively; is the attenuation factor; is the maximum number of iterations; is a random number obeying a normal distribution; is a random number uniformly distributed between 0 and 1; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1).

[0232] Example 3:

[0233] Referring to Figure 4 , a fixed-wing UAV swarm inspection mission planning and optimization device based on ICBBA-IGJO includes a memory and a processor; the memory is used to store computer program code and transmit the computer program code to the processor; the processor is configured to execute the optimization method described in Example 1 according to the instructions in the computer program code.

[0234] Example 4:

[0235] A computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the optimization method described in Example 1 is implemented.

[0236] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0237] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0238] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0239] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0240] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation manners of the present invention. Any modification or equivalent substitution that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. An optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO, characterized in that: The optimization method includes: Using the improved consensus bundling algorithm to generate an optimal task allocation scheme for the fixed-wing UAV swarm; Taking the minimum total inspection cost of the fixed-wing UAV swarm as the goal, constructing an optimization model for the inspection path of the UAV swarm; Based on the optimal task allocation scheme, using the improved golden jackal optimization algorithm to solve the optimization model for the inspection path of the UAV swarm to obtain the optimal inspection path.

2. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 1, characterized in that: The improved consensus bundling algorithm includes: S1, task package construction phase: construct the data structure required by the algorithm for the drones in the fixed-wing drone group, the data structure includes the Mission package for drones , Execution path list , Winners List , Winning Price List , the task package Indicates that the A collection of drone missions. The tasks are arranged in the order in which they are added, and the execution path list Indicates The mission execution path of the UAV, The tasks are arranged in the order of task execution, and the winner list Indicates storage The drone serial number set that the drone believes won the mission, the winning price list Indicates The set of bid prices of the drones that the drones believe will win the mission; S2. Task Selection Phase: Divide the total task set into an effective task set and an ineffective task set . With the goal of maximizing the total benefit of the fixed-wing UAV swarm, select tasks from the effective task set and add them to the task package ; S3. Redundancy Removal Phase: Perform validity checks on each task in the winner list ; if a task fails the validity check, it means that the task cannot reach a consensus through the preset conflict resolution rules, and the task is classified into the invalid task set ; if a task passes the validity check, it means that a consensus can be reached through the conflict resolution rules. Add the UAV serial number corresponding to this task in the winning price list to the valid winner list , and add the UAV bid price corresponding to this task in the winning price list to the valid winner price list , and then enter S4; perform the validity check according to the following formula: ； In the above formula, and are both indicator functions; is used to determine whether the winner or winning price of the th drone and the th drone adjacent to it in the communication topology can reach a local consensus on the task through the conflict resolution rule; is used to determine whether the winner or winning price of all drones and the drones adjacent to them in the communication topology can reach a local consensus on the task through the conflict resolution rule; S4. Conflict Resolution Phase: At step When a drone receives communication data sent by an adjacent drone in its communication topology, it updates its own timestamp. And based on the valid winner list and the valid winner price and the timestamp perform update, reset, replacement propagation or default operation through conflict resolution rules, and then update the data structure; S5. Return to step S2 for iterative calculation until the total benefit of the fixed-wing UAV swarm reaches the maximum.

3. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 2, characterized in that: The total benefit of the fixed-wing UAV swarm is the sum of the benefits of all UAVs, and the functional expression of the benefit of the UAV is: ； ； In the above formula, is the benefit of the th drone; is the th drone's time required to reach the task according to the execution path list ; is the benefit value of the th drone executing the task in the execution path list ; is the discount factor of the task in the execution path list ; is the clustering benefit of the execution path list ; and are the clustering categories of points and respectively; is the Euclidean distance between points and ; is the Euclidean distance between the clustering center to which point belongs and the clustering center to which point belongs; Point represents the new task added from the task package to the execution path list, and point represents the last task in the task package or the drone slot; and are the gain constant and the decay constant respectively.​ 4. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 1, characterized in that: The total inspection cost of the fixed-wing UAV swarm includes the total flight length cost, average flight height cost, total time consumption cost, and total power consumption cost of the fixed-wing UAV swarm. The calculation formula for the total inspection cost of the fixed-wing UAV swarm is: ； ； ； ； ； ； ； In the above formula, is the total inspection cost of the fixed-wing UAV swarm; is the total flight length cost of the fixed-wing UAV swarm; is the average flight altitude cost of the fixed-wing UAV swarm; is the total time consumption cost of the fixed-wing UAV swarm; is the total power consumption cost of the fixed-wing UAV swarm; , , , are the weight coefficients of the corresponding costs respectively; is the th UAV, , is the total number of UAVs in the fixed-wing UAV swarm; is the flight path from mission to mission ; is the Euclidean distance from mission to mission ; is the number of path points in path ; is a binary decision variable, which takes the value of 1 when both mission and mission belong to the mission sequence of UAV , otherwise takes the value of 0; , respectively represent the Z-axis coordinates of the th and th path points in path ; is the distance conversion coefficient; is the total flight distance of UAV ; is the number of missions assigned to UAV after the mission assignment; is the total flight time of UAV ; is the total number of missions of a single UAV; is the power consumption of UAV ; is the power per unit distance of the UAV. The constraint conditions of the optimization model for the inspection path of the UAV swarm include: Safety distance constraint: ； In the above formula, is the th unmanned aerial vehicle, , represents the distance between the unmanned aerial vehicle and the unmanned aerial vehicle or the distance between the unmanned aerial vehicle and the obstacle ; is the distance safety threshold.

5. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 1, characterized in that: The improved golden jackal optimization algorithm includes: first, determining the positions of male and female jackals according to the fitness values, where the male jackal is the jackal with the highest fitness value in the golden jackal population, and the female jackal is the jackal with the second highest fitness value in the golden jackal population; then, determining the position update strategy according to the prey escape energy If the prey escape energy satisfies , then the exploration strategy is adopted to update the relative positions of all jackal individuals and the prey, otherwise the exploitation strategy is adopted to update the relative positions of all jackal individuals and the prey; The escape energy of the prey The calculation formula is as follows: ； ； ； In the above formula, and are the escape random factor and the iteration number decay factor respectively; is a random value in (0, 1); is a constant; is the current iteration number; is the maximum iteration number.

6. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 5, characterized in that: The relative position update formula in the exploration strategy is: ； ； ； ； In the above formula, and respectively represent the relative position vectors of the male jackal and the female jackal with respect to the prey at the -th iteration; and respectively represent the position vectors of the male jackal and the female jackal at the -th iteration; represents the position vector of the prey at the -th iteration; represents the escape coefficient of the prey; represents a random number vector based on the Lévy distribution; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1).

7. The optimization method for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO according to claim 5, characterized in that: Using the golden jackal observation strategy to improve the relative position update formula of the exploitation strategy, and the improved relative position update formula of the exploitation strategy is: ； ； ； ； ； ； ； In the above formula, and respectively represent the relative position vectors of the male and female jackals with respect to the prey at the -th iteration; and respectively represent the position vectors of the male and female jackals at the -th iteration; and respectively represent the position vectors of the male and female jackals after observation; and are the upper and lower bounds of the position respectively; is the attenuation factor; is the maximum number of iterations; is a random number subject to a normal distribution; is a random number uniformly distributed between 0 and 1; is the relative position vector of the golden jackal individual with respect to the prey at the -th iteration; is the proportional balance factor; is a constant; is a random value in (0, 1).

8. An optimization system for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO, characterized in that: The optimization system includes a task allocation module, a model construction module, and a path optimization module; The task allocation module is used to generate an optimal task allocation scheme for the fixed-wing UAV swarm using the improved consensus bundling algorithm; The model construction module is used to construct an optimization model for the inspection path of the UAV swarm with the goal of minimizing the total inspection cost of the fixed-wing UAV swarm; The path optimization module is used to solve the optimization model for the inspection path of the UAV swarm based on the optimal task allocation scheme using the improved golden jackal optimization algorithm to obtain the optimal inspection path.

9. An optimization device for the inspection mission planning of a fixed-wing UAV swarm based on ICBBA-IGJO, characterized in that: The optimization device includes a memory and a processor; the memory is used for storing computer program code and transmitting the computer program code to the processor; the processor is used for executing the optimization method as described in claims 1-7 according to the instructions in the computer program code.

10. A computer-readable storage medium, characterized in that: A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the optimization method as described in claims 1-7 is implemented.

Images