Heterogeneous fixed wing cluster target distribution and planning method facing flight time coordination

Abstract

The invention belongs to the technical field of aircraft target allocation and planning, and particularly relates to a heterogeneous fixed wing cluster target allocation and planning method oriented to flight time collaboration, which comprises the following steps: firstly, introducing a KMEANS method, and clustering targets based on target positions to form target clusters corresponding to unmanned aerial vehicles; then, initial planning of the route in each target cluster is realized based on an improved genetic algorithm; and finally, redistributing and replanning targets in each target cluster on the basis of target position characteristics aiming at the endurance constraint caused by mutual difference of the optimal cruise speeds, so as to realize further reduction of endurance and endurance collaboration. The time-of-flight cooperative target allocation and planning algorithm is low in time complexity, low in computing power requirement and easy to deploy, and can be widely applied to top task planning and target decision allocation tasks of various heterogeneous unmanned aerial vehicle clusters in the future.

Description

Technical Field

[0001] This invention belongs to the field of aircraft target allocation and planning technology, specifically relating to a target allocation and planning method for heterogeneous fixed-wing swarms oriented towards flight time coordination. This technology can be applied to top-level mission planning, target decision allocation, and flight path planning tasks of various heterogeneous UAV swarms. Background Technology Unmanned aerial vehicle (UAV) swarms are playing an increasingly important role on the modern battlefield, and these swarms are increasingly evolving into heterogeneous clusters, as single-configuration swarms can no longer meet operational requirements. In multi-target close-in reconnaissance missions involving swarms, existing target allocation algorithms only allocate targets based on the characteristics of the target group, without considering the speed constraints caused by the heterogeneity of the UAVs. Furthermore, for each assigned target cluster, existing flight path planning algorithms are computationally inefficient when dealing with large-scale target sequence optimization problems. The varying flight times required for UAV swarms to complete target approach missions can hinder subsequent missions.

[0002] For flight time constraints in multi-target close-in reconnaissance missions involving heterogeneous UAV swarms, traditional target allocation and route planning methods do not consider flight time heterogeneity caused by speed heterogeneity, and cannot meet the flight time coordination requirements under battlefield conditions. Summary of the Invention

[0003] (a) Technical problems to be solved The technical problem this invention aims to solve is that traditional target allocation and flight path planning methods fail to consider the time-varying nature caused by speed heterogeneity and cannot meet the time-varying coordination requirements under battlefield conditions. Therefore, it is necessary to study target redistribution and replanning methods for heterogeneous UAV swarms. This invention provides a target allocation and planning technology for heterogeneous fixed-wing swarms oriented towards time-varying coordination.

[0004] (II) Technical Solution To address the aforementioned technical problems, this invention provides a method for target allocation and planning in heterogeneous fixed-wing formations oriented towards time-space coordination. The method includes the following steps: Step 1, Initial target allocation; Step two, initial route planning; Step 3, target reallocation; Step four: Overall target reallocation and route replanning.

[0005] In step one, the initial target allocation is performed. First, to quickly partition the targets, the k-means clustering algorithm is used, based on the distribution of the targets. and the number of available drones To initialize and divide the target clusters for each UAV , obtain the target cluster quantity Corresponding waypoints The Multiple Traveling Salesman Problem (MTSP) is transformed into the Single Traveling Salesman Problem (TSP). To simplify the expression, the following definition is used: .

[0006] In step two, the initial route planning is performed. For drones target cluster An optimized genetic algorithm is used to initialize the waypoint approach order, i.e., optimize the route; at this time, the set and All are ordered sets.

[0007] To achieve rapid route generation and address the curse of dimensionality caused by excessive waypoints, an improved distributed genetic algorithm is employed; the optimization objective is to minimize flight time. Defined as: in, Representing drones To which target cluster Inner One target point; The improved algorithm adds a 2-opt optimization operation, which can discard obviously erroneous offspring and reduce ineffective genetic iterations. The core consists of three parts: selection operator, crossover operator, and genetic operator. In the selection operator, a portion of individuals with high optimization is selected based on roulette wheel selection. A portion of individuals are randomly generated. The last portion selects individuals generated in the input order, and the proportion of this portion gradually increases with the number of iterations. In the crossover operator, parent crossover is performed by random slicing to avoid conflict detection and improve algorithm efficiency. In the mutation operator, a random slice of a certain offspring is rearranged to achieve rapid mutation.

[0008] In real-world scenarios, due to the number of targets... and location All are random, the optimal cruising speed of the drone They are also different, which will lead to drones Flight time obtained from the first optimization The differences are significant, reducing task execution efficiency.

[0009] In step three, the target is redistributed; This step proposes an inter-cluster iterative optimization algorithm, which reduces flight time and promotes collaboration through route iteration, flight time prediction, and target exchange. Specifically, after obtaining each target cluster Then, the algorithm for the target reassignment process is implemented to ensure the adaptive adjustment of each target cluster; in step 1 of the target reassignment process, In step 2 of the target reassignment process, to ensure that the target clusters are effectively adjusted in each iteration, the cluster adjustment order needs to be sorted according to... The process iterates from longest to shortest; in step 4 of the target reassignment process, in order to reassign the target point... The choice of target cluster is to be minimized, and the definition is as follows. Finding sets External distance set center point Recent target and its set ,Right now: definition To find a set Inner distance The farthest point : definition To find the distance The most recent and unadjusted set center point To which the set belongs ; exist In the middle, it is necessary to include Join an ordered set In practice, if the improved genetic algorithm is used to generate routes after each waypoint adjustment, the algorithm's efficiency will be reduced; in order to balance computational efficiency and optimization objectives, Finding distance nearest point and the corresponding next point ,Will Insert two points to ensure the target cluster and Still in order; in the target reallocation process, This represents the number of iterations.

[0010] The specific target reallocation process is as follows: Input: The initial and optimized set of target points target location set Number of drones drone cruising speed set ; Output: Optimized ordered set of waypoints ; Step 1: ; Calculate the center position of each target cluster and the average flight time of the UAV; Step 2: Define the longest flight time cluster The index is 1, and the results are recursively calculated in descending order of flight time until... ; Step 3: Loop ; Step 4: Loop and ; if otherwise Calculate the flight time of each UAV And update the average flight time ; End the loop.

[0011] In step four, the overall target is reassigned and the flight path is replanned. Following the overall process of target reallocation and route replanning, steps one through three are combined to achieve flight time coordination through continuous reallocation and replanning.

[0012] In step four, the overall process of target reassignment and route replanning is as follows: Input: Number of drones Cruise speed set Target location ; Output: Optimized ordered set of waypoints , ; Step A: Initialization; ; cycle ; ; End the loop; Step B: Loop and ; ; cycle ; ; ; End the loop; Here, The iteration count; ultimately ensuring balanced flight times for all drones while reducing average flight time. .

[0013] The method addresses the differences in flight time caused by the varying optimal cruising speeds of different UAVs in the cluster. Based on the target location characteristics, it redistributes and replans targets within each target cluster to achieve synchronous reduction and coordinated flight time.

[0014] (III) Beneficial Effects Compared with existing technologies, this invention addresses the aforementioned challenges by first introducing the K_MEANS method to perform initial clustering of targets based on their location information, forming target clusters corresponding to each UAV. Subsequently, an improved genetic algorithm is used to complete the initial flight path planning within each target cluster. Finally, to address the flight time constraints caused by differing optimal cruising speeds, specific targets are selected based on their location characteristics for inter-cluster redistribution and flight path replanning, ultimately achieving a further reduction in flight time while satisfying flight time coordination constraints. This flight time coordination target allocation and planning algorithm has low time complexity, low computational requirements, and is easy to deploy.

[0015] This technology is designed for multi-target close-range reconnaissance scenarios. The target allocation and flight path planning results obtained ensure the flight time coordination of heterogeneous UAV swarms. Moreover, the algorithm has low complexity and has broad application prospects. Attached Figure Description

[0016] Figure 1 The result diagram of target allocation and route initialization.

[0017] Figure 2 The result diagram after target allocation and route optimization.

[0018] Figure 3 The diagram illustrates the iterative process of optimizing flight distance and flight time.

[0019] Figure 4 This is a flowchart illustrating the overall technical solution of the present invention. Detailed Implementation

[0020] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0021] To address the aforementioned technical problems, this invention provides a method for target allocation and planning in heterogeneous fixed-wing formations oriented towards time-space coordination. The method includes the following steps: Step 1, Initial target allocation; Step two, initial route planning; Step 3, target reallocation; Step four: Overall target reallocation and route replanning.

[0022] In step one, the initial target allocation is performed. First, to quickly partition the targets, the k-means clustering algorithm is used, based on the distribution of the targets. and the number of available drones To initialize and divide the target clusters for each UAV , obtain the target cluster quantity Corresponding waypoints The Multiple Traveling Salesman Problem (MTSP) is transformed into the Single Traveling Salesman Problem (TSP). To simplify the expression, the following definition is used: .

[0023] In step two, the initial route planning is performed. For drones target cluster An optimized genetic algorithm is used to initialize the waypoint approach order, i.e., optimize the route; at this time, the set and All are ordered sets.

[0024] To achieve rapid route generation and address the curse of dimensionality caused by excessive waypoints, an improved distributed genetic algorithm is employed; the optimization objective is to minimize flight time. Defined as: in, Representing drones To which target cluster Inner One target point; The improved algorithm adds a 2-opt optimization operation, which can discard obviously erroneous offspring and reduce ineffective genetic iterations. The core consists of three parts: selection operator, crossover operator, and genetic operator. In the selection operator, a portion of individuals with high optimization is selected based on roulette wheel selection. A portion of individuals are randomly generated. The last portion selects individuals generated in the input order, and the proportion of this portion gradually increases with the number of iterations. In the crossover operator, parent crossover is performed by random slicing to avoid conflict detection and improve algorithm efficiency. In the mutation operator, a random slice of a certain offspring is rearranged to achieve rapid mutation.

[0025] In real-world scenarios, due to the number of targets... and location All are random, the optimal cruising speed of the drone They are also different, which will lead to drones Flight time obtained from the first optimization The differences are significant, reducing task execution efficiency.

[0026] In step three, the target is redistributed; This step proposes an inter-cluster iterative optimization algorithm, which reduces flight time and promotes collaboration through route iteration, flight time prediction, and target exchange. Specifically, after obtaining each target cluster Then, the algorithm for the target reassignment process is implemented to ensure the adaptive adjustment of each target cluster; in step 1 of the target reassignment process, In step 2 of the target reassignment process, to ensure that the target clusters are effectively adjusted in each iteration, the cluster adjustment order needs to be sorted according to... The process iterates from longest to shortest; in step 4 of the target reassignment process, in order to reassign the target point... The choice of target cluster is to be minimized, and the definition is as follows. Finding sets External distance set center point Recent target and its set ,Right now: definition To find a set Inner distance The farthest point : definition To find the distance The most recent and unadjusted set center point To which the set belongs ; exist In the middle, it is necessary to include Join an ordered set In practice, if the improved genetic algorithm is used to generate routes after each waypoint adjustment, the algorithm's efficiency will be reduced; in order to balance computational efficiency and optimization objectives, Finding distance nearest point and the corresponding next point ,Will Insert two points to ensure the target cluster and Still in order; in the target reallocation process, This represents the number of iterations.

[0027] The specific target reallocation process is as follows: Input: The initial and optimized set of target points target location set Number of drones drone cruising speed set ; Output: Optimized ordered set of waypoints ; Step 1: ; Calculate the center position of each target cluster and the average flight time of the UAV; Step 2: Define the longest flight time cluster The index is 1, and the results are recursively calculated in descending order of flight time until... ; Step 3: Loop ; Step 4: Loop and ; if otherwise Calculate the flight time of each UAV And update the average flight time ; End the loop.

[0028] In step four, the overall target is reassigned and the flight path is replanned. Following the overall process of target reallocation and route replanning, steps one through three are combined to achieve flight time coordination through continuous reallocation and replanning.

[0029] In step four, the overall process of target reassignment and route replanning is as follows: Input: Number of drones Cruise speed set Target location ; Output: Optimized ordered set of waypoints , ; Step A: Initialization; ; cycle ; ; End the loop; Step B: Loop and ; ; cycle ; ; ; End the loop; Here, The iteration count; ultimately ensuring balanced flight times for all drones while reducing average flight time. .

[0030] The method addresses the differences in flight time caused by the varying optimal cruising speeds of different UAVs in the cluster. Based on the target location characteristics, it redistributes and replans targets within each target cluster to achieve synchronous reduction and coordinated flight time.

[0031] Example 1 This embodiment further describes the invention using a two-dimensional computational example of a fixed-wing swarm. The swarm's starting and ending points are both the origin [0m, 0m], and the swarm size and target location set are also specified. exist Randomly generated, number of drones drone cruising speed set . Step 1, Initial Target Allocation The k-means clustering algorithm is used, based on the distribution of the target. and the number of available drones Initialize the division of target clusters for each UAV , obtain the target cluster quantity and corresponding waypoints .

[0032] Step 2, Initial Route Planning For drones target cluster The optimized flight path is initialized based on the optimized genetic algorithm.

[0033] Initialized target allocation and flight path are as follows: Figure 1 As shown, the curves with different line shapes represent the flight paths of different UAVs. It can be seen that the number of targets in each target cluster is basically the same, but the flight time varies greatly depending on the different cruising speeds.

[0034] Step 3, Target Reassignment Implement the target redistribution algorithm in Table 1, adaptively adjusting the targets in each target cluster. In step 1 of Table 1... In step 2, to ensure that the target cluster is effectively adjusted in each iteration, the cluster adjustment order needs to be sorted according to... Iterate from longest to shortest. .

[0035] Step 4: Overall target reallocation and route replanning Execute the overall process for target reallocation and route replanning as shown in Table 2. This ensures balanced flight times for all drones while also reducing the average flight time. .

[0036] Table 1 Target Reassignment Process Table 2 Overall Process of Target Reassignment and Route Replanning To verify the advantages of this scheme, a comparison of the route and flight time before and after optimization was conducted, and digital simulation was used to obtain the results. Figures 2-3 The curve. Figure 2 The allocation and order of targets are shown, with each target's location represented by solid dots, light colors indicating the target was detected first, and dark colors the next; different line types represent different UAV flight paths. Among these, due to... Its cruising speed is the highest, so its optimized range is the longest; Secondly, The shortest route. (By...) Figure 3 As optimization progresses, the flight times gradually become more consistent, with the flight times of the four drones being as follows: The average flight time decreased from 218s to 184s.

[0037] This invention achieves coordinated flight time after target reassignment and route replanning, and reduces the average flight time of UAVs, ensuring the rapid and effective completion of coordinated close-range reconnaissance missions.

[0038] In summary, this invention belongs to the field of aircraft target allocation and planning technology, specifically involving a heterogeneous fixed-wing swarm target allocation and planning method for time-coordination. Based on target redistribution and replanning among swarms, it rapidly reconstructs and optimizes target clusters in each swarm, effectively ensuring that fixed-wing UAVs with different optimal cruise speeds can simultaneously complete target reconnaissance missions. For scenarios where heterogeneous fixed-wing UAVs perform multi-target close-in reconnaissance missions, existing target allocation and route optimization algorithms do not consider the constraint of different optimal cruise speeds among the heterogeneous UAVs, resulting in different completion times for reconnaissance sub-tasks by the UAV swarm. To address this problem, this invention first introduces the K_MEANS method to cluster targets based on their locations, forming target clusters corresponding to each UAV. Then, an improved genetic algorithm is used to initially plan the routes within each target cluster. Finally, to address the time constraints caused by differing optimal cruise speeds, targets within each target cluster are redistributed and replanned based on target location characteristics, further reducing time and achieving time-coordination. This time-coordinated target allocation and planning algorithm has low time complexity, low computing power requirements, and is easy to deploy. In the future, it can be widely used in top-level mission planning and target decision allocation tasks for various heterogeneous UAV swarms.

[0039] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for target allocation and planning in heterogeneous fixed-wing formations oriented towards time-space coordination, characterized in that, The method includes the following steps: Step 1, Initial target allocation; Step two, initial route planning; Step 3, target reallocation; Step four: Overall target reallocation and route replanning.

2. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards flight time coordination as described in claim 1, characterized in that, In step one, the initial target allocation; First, to quickly partition the targets, the k-means clustering algorithm is used, based on the distribution of the targets. and the number of available drones To initialize and divide the target clusters for each UAV , obtain the target cluster quantity Corresponding waypoints The Multiple Traveling Salesman Problem (MTSP) is transformed into the Single Traveling Salesman Problem (TSP). To simplify the expression, the following definition is used: .

3. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards flight time coordination as described in claim 2, characterized in that, In step two, the initial route planning is performed. For drones target cluster An optimized genetic algorithm is used to initialize the waypoint approach order, i.e., optimize the route; at this time, the set and All are ordered sets.

4. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards flight time coordination as described in claim 3, characterized in that, To achieve rapid route generation and address the curse of dimensionality caused by excessive waypoints, an improved distributed genetic algorithm is employed; the optimization objective is to minimize flight time. Defined as: in, Representing drones To which target cluster Inner One target point; The improved algorithm adds a 2-opt optimization operation, which can discard obviously erroneous offspring and reduce ineffective genetic iterations. The core consists of three parts: selection operator, crossover operator, and genetic operator. In the selection operator, a portion of individuals with high optimization is selected based on roulette wheel selection. A portion of individuals are randomly generated. The last portion selects individuals generated in the input order, and the proportion of this portion gradually increases with the number of iterations. In the crossover operator, parent crossover is performed by random slicing to avoid conflict detection and improve algorithm efficiency. In the mutation operator, a random slice of a certain offspring is rearranged to achieve rapid mutation.

5. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 4, characterized in that, In real-world scenarios, due to the number of targets... and location All are random, the optimal cruising speed of the drone They are also different, which will lead to drones Flight time obtained from the first optimization The differences are significant, reducing task execution efficiency.

6. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 5, characterized in that, In step three, the target is reassigned; This step proposes an inter-cluster iterative optimization algorithm, which reduces flight time and promotes collaboration through route iteration, flight time prediction, and target exchange. Specifically, after obtaining each target cluster Then, the algorithm for the target reassignment process is implemented to ensure the adaptive adjustment of each target cluster; in step 1 of the target reassignment process, In step 2 of the target reassignment process, to ensure that the target clusters are effectively adjusted in each iteration, the cluster adjustment order needs to be sorted according to... The process iterates from longest to shortest; in step 4 of the target reassignment process, in order to reassign the target point... The choice of target cluster is to be minimized, and the definition is as follows. Finding sets External distance set center point Recent target and its set ,Right now: definition To find a set Inner distance The farthest point : definition To find the distance The most recent and unadjusted set center point To which the set belongs ; exist In the middle, it is necessary to include Join an ordered set In practice, if the improved genetic algorithm is used to generate routes after each waypoint adjustment, the algorithm's efficiency will be reduced; in order to balance computational efficiency and optimization objectives, Finding distance nearest point and the corresponding next point ,Will Insert two points to ensure the target cluster and Still in order; in the target reallocation process, This represents the number of iterations.

7. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 6, characterized in that, The target reallocation process is as follows: Input: The initial and optimized set of target points target location set Number of drones drone cruising speed set ; Output: Optimized set of ordered waypoints ; Step 1: ; Calculate the center position of each target cluster and the average flight time of the UAV; Step 2: Define the longest flight time cluster The index is 1, and the results are recursively calculated in descending order of flight time until... ; Step 3: Loop ; Step 4: Loop and ; if otherwise Calculate the flight time of each UAV And update the average flight time ； End the loop.

8. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 7, characterized in that, In step four, the overall target is reassigned and the flight routes are replanned. Following the overall process of target reallocation and route replanning, steps one through three are combined to achieve flight time coordination through continuous reallocation and replanning.

9. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 8, characterized in that, In step four, the overall process of target reallocation and route replanning is as follows: Input: Number of drones Cruise speed set Target location ; Output: Optimized set of ordered waypoints , ; Step A: Initialization; ； cycle ; ； End the loop; Step B: Loop and ; ； cycle ; ； ； End the loop; Here, The iteration count; ultimately ensuring balanced flight times for all drones while reducing average flight time. .

10. The method for target allocation and planning of heterogeneous fixed-wing group oriented towards time-space coordination as described in claim 9, characterized in that, The method addresses the time differences caused by the varying optimal cruising speeds of different UAVs in the cluster by redistributing and replanning targets within each target cluster based on target location characteristics, thereby achieving synchronous reduction and coordinated time management.

Images