A multi-unmanned aerial vehicle task allocation method, device, equipment and storage medium

Abstract

The present invention provides a multi-UAV task allocation method, device, equipment and storage medium, belonging to the technical field of UAV task allocation. The method initializes task target data and UAV cluster data, and calculates the cost information of each UAV in the UAV cluster to execute each task target; based on the cost information of each task target, a dynamic programming strategy is used to select the optimal task target set to be executed; a greedy strategy is used to select the UAV with the lowest cost to execute each optimal task target from the UAV cluster in turn for task allocation, until the task allocation of the optimal task target set is completed. The present invention can maximize the task value, reduce the cost of UAV execution from the local optimum, and maximize the execution success rate of the current task.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) task allocation, and in particular to a multi-UAV task allocation method, device, equipment and storage medium. Background Art

[0002] UAV task allocation is the key technology for UAV swarms to achieve efficient combat missions. With the development of UAV swarm technology and the change of combat style, the combat mission field of UAV swarms is constantly expanding, the scope of task allocation is constantly expanding, and the scale and complexity of task allocation problems are constantly increasing, which poses new challenges to UAV swarm task allocation technology.

[0003] The traditional multi-task drone allocation algorithm mainly considers assigning an unassigned drone to a task, but drones face complex environments and have failure problems. When the mission target is of great value to drones, in order to ensure the success rate of mission execution, it is necessary to assign a task to a recommended number of drones, and have multiple drones execute the task at the same time to ensure the success rate of mission execution as much as possible. When the value of each task is different, the number of drones required is different, and the number of drones is limited, it is necessary to solve the problem of how to allocate to maximize the value. At the same time, when the cost of drones executing each task is different, it is necessary to solve how to reduce the cost of unmanned missions. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a multi-UAV task allocation method, device, equipment and storage medium to solve the problem that the existing UAV allocation method is difficult to maximize the task value while reducing the cost of UAV task execution.

[0005] The technical solution adopted by the present invention is:

[0006] In a first aspect, the present invention provides a multi-UAV task allocation method, comprising:

[0007] Initialize the mission target data and drone cluster data, and calculate the cost information of each drone in the drone cluster to execute each mission target;

[0008] Based on the cost information of each task goal, a dynamic programming strategy is used to select the optimal set of task goals to be executed;

[0009] A greedy strategy is adopted to select the drone with the lowest cost to execute each optimal mission objective from the drone cluster for task assignment until the task assignment of the optimal mission objective set is completed.

[0010] Furthermore, based on the cost information of each task goal, a dynamic programming strategy is used to select the optimal task goal set to be executed, including:

[0011] S1: Initialize and set the number of mission targets to N, the number of drones required for the mission target to c[N], the mission target value to v[N], the total number of drones to C, the maximum value array to dp[N+1][C+1], dp[N+1][C+1]=0, and the mission target selection array to a[N], a[N]=0;

[0012] S2: Initialize task number i=1;

[0013] S3: Determine whether i≤N is satisfied. If so, continue to execute; if not, enter S6;

[0014] S3: Initialize the number of drones j=1;

[0015] S4: Determine whether j≤C is satisfied. If so, continue to execute; if not, set i=i+1 and return to execute S3;

[0016] S5: Determine whether the number of drones required for the current mission goal satisfies c[i]>j. If so, set dp[i][j]=dp[i-1][j], j=j+1, and return to execute S4; if not, calculate the maximum value array Max(dp[i-1][j], dp[i-1][jj[i]]+v[i]), and set j=j+1, and return to execute S4;

[0017] S6: Let the mission number i=N, the number of drones j=C;

[0018] S7: Determine whether i≥1 is satisfied. If so, proceed to S8; if not, output the target selection array a[ ], where the element 1 represents the selected task target;

[0019] S8: Determine whether dp[i][j]=dp[i-1][j] is satisfied. If so, select the task target a[i]=0, set i=i-1, and return to execute S7; if not, select the task target a[i]=1, j=jc[i], set i=i-1, and return to execute S7;

[0020] S9: Use the output target selection array a[ ] as the optimal task target set to be executed in the end.

[0021] Furthermore, the greedy strategy is used to select the drone with the lowest cost to execute each optimal mission target from the drone cluster in turn for task allocation, including:

[0022] Select the selected but unassigned UAV mission targets from the optimal mission target set in turn;

[0023] Calculate the execution cost of all unassigned drones in the drone cluster to execute the current mission objectives;

[0024] Assign the drone with the lowest execution cost to the current mission target until the number of drones required for the current mission target is met. Record the task number of the current mission target and the corresponding assigned drone number, and repeatedly select the mission targets with unassigned drones for drone assignment until all mission targets in the optimal mission target set are assigned.

[0025] In a second aspect, the present invention provides a multi-UAV task allocation device, comprising:

[0026] The mission cost calculation module is used to initialize the mission target data and the UAV cluster data, and calculate the cost information of each UAV in the UAV cluster to execute each mission target;

[0027] The target screening module is used to select the optimal set of task targets to be executed based on the cost information of each task target using a dynamic programming strategy;

[0028] The task allocation module is used to adopt a greedy strategy to select the drone with the lowest cost to execute each optimal task target from the drone cluster for task allocation until the task allocation of the optimal task target set is completed.

[0029] Furthermore, the task allocation module specifically includes:

[0030] A task traversal unit, used for sequentially selecting task targets that are selected but not assigned to drones from the optimal task target set;

[0031] An execution cost calculation unit is used to calculate the execution cost of all unassigned drones in the drone cluster to execute the current task target;

[0032] The drone task allocation unit is used to allocate the drone with the lowest execution cost to the current task target until the number of drones required for the current task target is met, record the task number of the current task target and the corresponding allocated drone number, and repeatedly select task targets to which drones are not allocated for drone allocation until all task targets in the optimal task target set are allocated.

[0033] In a third aspect, the present invention provides an electronic device comprising: at least one processor, at least one memory, and computer program instructions stored in the memory, wherein the method of the first aspect is implemented when the computer program instructions are executed by the processor.

[0034] In a fourth aspect, the present invention provides a storage medium having computer program instructions stored thereon, which implement the method of the first aspect when the computer program instructions are executed by a processor.

[0035] In summary, the beneficial effects of the present invention are as follows:

[0036] The method for allocating tasks among multiple UAVs provided by the present invention first initializes the task target data and the UAV cluster data, and calculates the cost information of each UAV in the UAV cluster to execute each task target. Then, based on the cost information of each task target, a dynamic programming strategy is used to select the optimal task target set to be executed, and the task target combination with the greatest value is obtained to maximize the task value. Finally, a greedy strategy is used to select the UAVs with the lowest cost to execute each optimal task target from the UAV cluster in turn for task allocation, until the task allocation of the optimal task target set is completed, thereby reducing the cost of UAV execution from the local optimum and maximizing the execution success rate of the current task. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the following is a brief introduction to the drawings required for use in the embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work, and these are all within the protection scope of the present invention.

[0038] Figure 1 This is a flow chart of a multi-UAV task allocation method of the present invention;

[0039] Figure 2 This is a schematic diagram of the optimal target set selection process of the present invention;

[0040] Figure 3 This is a flow chart of the method for allocating drones in the optimal target set of the present invention;

[0041] Figure 4 A schematic diagram of a UAV allocation process in an application scenario of the present invention;

[0042] Figure 5 This is a functional module diagram of a multi-UAV task allocation device of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. If there is no conflict, the various features of the present invention and the embodiments can be combined with each other, all within the scope of protection of the present invention.

[0044] Example 1: Please see Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for allocating tasks to multiple drones in Embodiment 1 of the present invention. Figure 1 As shown, the present invention provides a multi-UAV task allocation method, comprising:

[0045] Initialize the mission target data and drone cluster data, and calculate the cost information of each drone in the drone cluster to execute each mission target; the mission target data includes the mission number of each mission, the value of the mission target, the number of drones required for the mission, and the cost information of the current targets of other drones. The drone cluster data includes the number of each drone.

[0046] Based on the cost information of each task objective, a dynamic programming strategy is used to select the optimal set of task objectives to be executed.

[0047] A greedy strategy is adopted to select the drone with the lowest cost to execute each optimal mission objective from the drone cluster for task assignment until the task assignment of the optimal mission objective set is completed.

[0048] Specifically, in some embodiments, the drone cluster needs to search for targets on the map. Then, the drone cluster generates tasks for the searched mission targets and calculates the cost of executing the mission targets. Tasks are assigned according to the cost value, and the drone cluster executes the designed path planning and executes the tasks according to the task assignment results. Specifically, when calculating the cost of executing the mission target, the drone cluster can calculate it through the cost function. Among them, the factors affecting the cost of the mission target are the total flight time, the total flight distance and the mission target value. Therefore, the cost function can be set as:

[0049] dp = αt+βL+vP

[0050] in, dp represents the cost of executing the task goal, t Indicates the total time to execute the task. L Indicates the total distance of the estimated flight. v Indicates the value of the task goal; α , β , v is the weight of each influencing factor.

[0051] In an embodiment of the present invention, the multi-unmanned task allocation process for multiple task targets is mainly processed in two stages. In the first stage, a dynamic programming strategy is used to select the task target combination scheme with the greatest value to obtain the optimal task target set; in the second stage, a greedy strategy is used for the task targets selected in the set, and a suitable drone is selected and allocated to each task target according to the cost of the drone to perform the task. Compared with the traditional allocation method in which only one drone is allocated to each task, the allocation of a suitable number of drones according to the task requirements maximizes the execution success rate of the current task. At the same time, in the case of a limited number of drones, compared with the requirement of the number of drones that meet the task requirements in sequence, the present invention uses a dynamic programming algorithm to select the task target with a small task execution cost as the optimal target, thereby improving the task value and maximizing the task value. In addition, compared with assigning tasks to each drone in turn, the present invention uses a greedy algorithm to assign tasks to drones, which reduces the cost of drone execution from a local optimum.

[0052] Further, in the embodiment of the present invention, referring to Figure 2 As shown in the figure, based on the cost information of each task goal, a dynamic programming strategy is used to select the optimal set of task goals to be executed, which specifically includes the following execution steps:

[0053] S1: Initialize and set the number of mission targets to N, the number of drones required for the mission target to c[N], the mission target value to v[N], the total number of drones to C, the maximum value array to dp[N+1][C+1], dp[N+1][C+1]=0, and the mission target selection array to a[N], a[N]=0;

[0054] S2: Initialize task number i=1;

[0055] S3: Determine whether i≤N is satisfied. If so, continue to execute; if not, enter S6;

[0056] S3: Initialize the number of drones j=1;

[0057] S4: Determine whether j≤C is satisfied. If so, continue to execute; if not, set i=i+1 and return to execute S3;

[0058] S5: Determine whether the number of drones required for the current mission goal satisfies c[i]>j. If so, set dp[i][j]=dp[i-1][j], j=j+1, and return to execute S4; if not, calculate the maximum value array Max(dp[i-1][j], dp[i-1][jj[i]]+v[i]), set j=j+1, and return to execute S4;

[0059] S6: Let the mission number i=N, the number of drones j=C;

[0060] S7: Determine whether i≥1 is satisfied. If so, proceed to S8; if not, output the target selection array a[ ], where the element 1 represents the selected task target;

[0061] S8: Determine whether dp[i][j]=dp[i-1][j] is satisfied. If so, select the task target a[i]=0, set i=i-1, and return to execute S7; if not, select the task target a[i]=1, j=jc[i], set i=i-1, and return to execute S7;

[0062] S9: Use the output target selection array a[ ] as the optimal task target set to be executed in the end.

[0063] Among them, the dynamic programming algorithm is applicable to problems with overlapping subproblems and optimal substructure properties. Its basic idea is to decompose the problem to be solved into several related subproblems, solve the subproblems first, and then use the solutions of these subproblems to construct the solution of the original problem. Therefore, the present invention adopts a dynamic programming strategy based on the dynamic programming algorithm to select the optimal task goal set from multiple task goals to maximize the task value.

[0064] Further, in the embodiment of the present invention, referring to Figure 3 As shown in the figure, a greedy strategy is adopted to select the drone with the lowest cost to execute each optimal mission target from the drone cluster for task allocation, including:

[0065] Select the selected but unassigned UAV mission targets from the optimal mission target set in turn;

[0066] Calculate the execution cost of all unassigned drones in the drone cluster to execute the current mission objectives;

[0067] Assign the drone with the lowest execution cost to the current mission target until the number of drones required for the current mission target is met. Record the task number of the current mission target and the corresponding assigned drone number, and repeatedly select the mission targets with unassigned drones for drone assignment until all mission targets in the optimal mission target set are assigned.

[0068] Specifically, a greedy algorithm is an algorithm strategy that takes the best or optimal choice in the current state at each step, hoping to lead to the best or optimal result in the world. A greedy algorithm does not guarantee the optimal solution, but in some problems, the solution of a greedy algorithm is close enough to the optimal solution or is indeed the optimal solution. The core idea of ​​a greedy algorithm is to always make the best choice at the moment when solving a problem, without considering the overall optimal solution, but to achieve the overall optimal solution through a series of local optimal choices.

[0069] When calculating the execution cost of all unassigned drones to execute the current mission target, the cost function described in the embodiment of the present invention may also be used, and the embodiment of the present invention will not be described in detail here.

[0070] The embodiment of the present invention adopts a greedy strategy based on a greedy algorithm to allocate a suitable drone to each mission target, thereby achieving local optimum in drone mission allocation and reducing the cost of drone mission execution from the local optimum.

[0071] Further, in some other embodiments, referring to Figure 4 As shown, the method of the embodiment of the present invention is applied to a battlefield environment to achieve strikes on multiple detected targets. Specifically, the value of each target and the number of drones recommended for strike are analyzed first, and then the drone task allocation algorithm (method) of the present invention is used to allocate tasks based on the existing strike drone cluster information, determine the strike target of each strike drone, and finally strike the strike target, thereby reducing the cost of drones performing tasks in an actual battlefield environment.

[0072] Example 2: Reference Figure 5 As shown, based on Example 1, the embodiment of the present invention further provides a multi-UAV task allocation device, including:

[0073] The mission cost calculation module is used to initialize the mission target data and the UAV cluster data, and calculate the cost information of each UAV in the UAV cluster to execute each mission target;

[0074] The target screening module is used to select the optimal set of task targets to be executed based on the cost information of each task target using a dynamic programming strategy;

[0075] The task allocation module is used to adopt a greedy strategy to select the drone with the lowest cost to execute each optimal task target from the drone cluster for task allocation until the task allocation of the optimal task target set is completed.

[0076] Furthermore, in the embodiment of the present invention, the task allocation module specifically includes:

[0077] A task traversal unit, used for sequentially selecting task targets that are selected but not assigned to drones from the optimal task target set;

[0078] An execution cost calculation unit is used to calculate the execution cost of all unassigned drones in the drone cluster to execute the current task target;

[0079] The drone task allocation unit is used to allocate the drone with the lowest execution cost to the current task target until the number of drones required for the current task target is met, record the task number of the current task target and the corresponding allocated drone number, and repeatedly select task targets to which drones are not allocated for drone allocation until all task targets in the optimal task target set are allocated.

[0080] Specifically, in the embodiment of the present invention, the process of selecting the optimal task target set by the target screening module is as follows: S1: Initialize and set the number of task targets as N, the number of drones required for the task target as c[N], the value of the task target as v[N], the total number of drones as C, the maximum value array as dp[N+1][C+1], dp[N+1][C+1]=0, and the array a[N] indicating whether the task target is selected, a[N]=0;

[0081] S2: Initialize task number i=1;

[0082] S3: Determine whether i≤N is satisfied. If so, continue to execute; if not, enter S6;

[0083] S3: Initialize the number of drones j=1;

[0084] S4: Determine whether j≤C is satisfied. If so, continue to execute; if not, set i=i+1 and return to execute S3;

[0085] S5: Determine whether the number of drones required for the current mission goal satisfies c[i]>j. If so, set dp[i][j]=dp[i-1][j], j=j+1, and return to execute S4. If not, calculate Max(dp[i-1][j],dp[i-1][jj[i]]+v[i]), set j=j+1, and return to execute S4.

[0086] S6: Let the mission number i=N, the number of drones j=C;

[0087] S7: Determine whether i≥1 is satisfied. If so, proceed to S8; if not, output the target selection array a[ ], where the element 1 represents the selected task target;

[0088] S8: Determine whether dp[i][j]=dp[i-1][j] is satisfied. If so, select the task target a[i]=0, set i=i-1, and return to execute S7; if not, select the task target a[i]=1, j=jc[i], set i=i-1, and return to execute S7;

[0089] S9: Use the output target selection array a[ ] as the optimal task target set to be executed in the end.

[0090] The embodiment of the present invention initializes the task target data and the drone cluster data, and calculates the cost information of each drone in the drone cluster to execute each task target. Then, based on the cost information of each task target, a dynamic programming strategy is used to select the optimal task target set to be executed, and the task target combination with the greatest value is obtained to maximize the task value. Finally, the greedy strategy is used to select the drones with the lowest cost to execute each optimal task target from the drone cluster for task allocation, until the task allocation of the optimal task target set is completed, thereby reducing the cost of drone execution from the local optimum and maximizing the execution success rate of the current task.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technologies therein with equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-UAV task allocation method, characterized in that: include: Initialize the mission target data and drone cluster data, and calculate the cost information of each drone in the drone cluster to execute each mission target; When the cost function is used to calculate the cost of executing the task objective, the cost function is set to: dp=αt+βL+vP Where dp represents the cost of executing the mission goal, t represents the total time of executing the mission, L represents the total distance expected to fly, and v represents the mission goal value; α, β, and v are the weights of each influencing factor; Based on the cost information of each task goal, a dynamic programming strategy is used to select the optimal set of task goals to be executed, including: S1: Initialize and set the number of mission targets to N, the number of drones required for the mission target to c[N], the mission target value to v[N], the total number of drones to C, the maximum value array to dp[N+1][C+1], dp[N+1][C+1]=0, and the mission target selection array to a[N], a[N]=0; S2: Initialize task number i=1; S3: Determine whether i≤N is satisfied. If so, continue to execute; if not, enter S6; S3: Initialize the number of drones j = 1; S4: Determine whether j≤C is satisfied. If so, continue to execute; if not, set i=i+1 and return to execute S3; S5: Determine whether the number of drones required for the current mission goal satisfies c[i]>j. If so, set dp[i][j]=dp[i-1][j], j=j+1, and return to execute S4; if not, calculate the maximum value array Max(dp[i-1][j], dp[i-1][jj[i]]+v[i]), set j=j+1, and return to execute S4; S6: Let the mission number i = N, and the number of drones j = C; S7: Determine whether i≥1 is satisfied. If so, proceed to S8; if not, output the target selection array a[], where the element 1 represents the selected task target; S8: Determine whether dp[i][j]=dp[i-1][j] is satisfied. If so, select task target a[i]=0, set i=i-1, and return to execute S7; if not, select task target a[i]=1, j=jc[i], set i=i-1, and return to execute S7; S9: Use the output target selection array a[] as the optimal task target set to be executed in the end; A greedy strategy is adopted to select the drone with the lowest cost to execute each optimal mission objective from the drone cluster for task assignment until the task assignment of the optimal mission objective set is completed.

2. The multi-UAV task allocation method according to claim 1, characterized in that: The greedy strategy is used to select the drone with the lowest cost to execute each optimal mission target from the drone cluster in turn for task allocation, including: Select the selected but unassigned UAV mission targets from the optimal mission target set in turn; Calculate the execution cost of all unassigned drones in the drone cluster to execute the current mission objectives; Assign the drone with the lowest execution cost to the current mission target until the number of drones required for the current mission target is met. Record the task number of the current mission target and the corresponding assigned drone number, and repeatedly select the mission targets with unassigned drones for drone assignment until all mission targets in the optimal mission target set are assigned.

3. A multi-UAV task allocation device, characterized in that: include: The mission cost calculation module is used to initialize the mission target data and the UAV cluster data, and calculate the cost information of each UAV in the UAV cluster to execute each mission target; When the cost function is used to calculate the cost of executing the task objective, the cost function is set to: dp=αt+βL+vP Where dp represents the cost of executing the mission goal, t represents the total time of executing the mission, L represents the total distance expected to fly, and v represents the mission goal value; α, β, and v are the weights of each influencing factor; The target screening module is used to select the optimal set of task targets to be executed based on the cost information of each task target using a dynamic programming strategy, including: S1: Initialize and set the number of mission targets to N, the number of drones required for the mission target to c[N], the mission target value to v[N], the total number of drones to C, the maximum value array to dp[N+1][C+1], dp[N+1][C+1]=0, and the mission target selection array to a[N], a[N]=0; S2: Initialize task number i=1; S3: Determine whether i≤N is satisfied. If so, continue to execute; if not, enter S6; S3: Initialize the number of drones j = 1; S4: Determine whether j≤C is satisfied. If so, continue to execute; if not, set i=i+1 and return to execute S3; S5: Determine whether the number of drones required for the current mission goal satisfies c[i]>j. If so, set dp[i][j]=dp[i-1][j], j=j+1, and return to execute S4; if not, calculate the maximum value array Max(dp[i-1][j], dp[i-1][jj[i]]+v[i]), set j=j+1, and return to execute S4; S6: Let the mission number i = N, and the number of drones j = C; S7: Determine whether i≥1 is satisfied. If so, proceed to S8; if not, output the target selection array a[], where the element 1 represents the selected task target; S8: Determine whether dp[i][j]=dp[i-1][j] is satisfied. If so, select task target a[i]=0, set i=i-1, and return to execute S7; if not, select task target a[i]=1, j=jc[i], set i=i-1, and return to execute S7; S9: Use the output target selection array a[] as the optimal task target set to be executed in the end; The task allocation module is used to adopt a greedy strategy to select the drone with the lowest cost to execute each optimal task target from the drone cluster for task allocation until the task allocation of the optimal task target set is completed.

4. The multi-UAV task allocation device according to claim 3, characterized in that: The task allocation module specifically includes: A task traversal unit, used for sequentially selecting task targets that are selected but not assigned to drones from the optimal task target set; An execution cost calculation unit is used to calculate the execution cost of all unassigned drones in the drone cluster to execute the current task target; The drone task allocation unit is used to allocate the drone with the lowest execution cost to the current task target until the number of drones required for the current task target is met, record the task number of the current task target and the corresponding allocated drone number, and repeatedly select task targets to which drones are not allocated for drone allocation until all task targets in the optimal task target set are allocated.

5. An electronic device, characterized in that: include: At least one processor, at least one memory and computer program instructions stored in the memory, when the computer program instructions are executed by the processor, implement the method according to any one of claims 1-2.

6. A storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 2 is implemented.

Images