An inter-group task resource coordination and allocation method for unmanned cluster command control

Abstract

The application discloses a task resource coordination and distribution method for unmanned cluster command control, and belongs to the technical field of unmanned cluster command control, and comprises the following steps: extracting all unmanned cluster tasks and corresponding requirements of each task, extracting task activity sequences of the tasks and resource requirements of each task activity in sequence based on respective meta-task models of the tasks, calculating feasible unmanned clusters of each single task based on the resource requirements and available task resource sets, obtaining a multi-task candidate unmanned cluster list, when there is a resource contention, realizing automatic coordination and distribution of the contended resources among unmanned clusters based on overall benefit maximization of formation tasks, and when there is no resource contention, selecting an unmanned cluster with the highest comprehensive evaluation value as a task resource distribution result. The application can realize a task resource distribution method suitable for unmanned aerial vehicle clusters, unmanned ship clusters and unmanned underwater vehicle clusters in a multi-domain and multi-task condition.

Description

Technical Field

[0001] This invention belongs to the field of unmanned swarm command and control technology, and more specifically, relates to a method for allocating task resources among unmanned swarm groups. Background Technology

[0002] Universality requires unmanned combat systems to be able to carry a sufficient number of mission payloads and adapt to the needs of various tactics. Integration refers to the development of unmanned combat systems towards multi-platform, clustered collaboration, and multi-system cooperation, enabling them to perform missions independently as well as conduct cooperative operations.

[0003] With the development trend of intelligentization, generalization, and integration of unmanned combat systems, a corresponding command and control capability is required. Among these, task resource allocation is one of the key issues to be addressed in the command and control of unmanned swarms. Existing research, in terms of content, focuses primarily on the coordination and allocation of task resources within an unmanned swarm; research results on inter-swarm task resource coordination and allocation are limited. Technically, optimization solutions are often found using genetic algorithms, particle swarm optimization, tabu search, simulated annealing, and market auction mechanisms. However, when there are many tasks, complex task activity sequences, numerous task constraints, and multiple combinations of task resources, problems arise such as the inability to achieve optimal solutions within a reasonable timeframe, the inability to guarantee compliance with various task scenarios, and the failure to fully consider task constraints, leading to resource combination errors or omissions. Therefore, an efficient and universal solution is still lacking for the coordination and allocation of inter-swarm task resources in multi-domain, multi-task swarm collaboration scenarios. Summary of the Invention

[0004] To address the potential mission conflicts arising from competition for unmanned swarm resources in multi-domain operations, as well as conflicts in mission allocation, this invention proposes a method for coordinating and allocating mission resources among unmanned swarms for command and control. This method is applicable to multi-domain, multi-mission scenarios for allocating mission resources among unmanned aerial vehicle (UAV) swarms, unmanned surface vessel (USV) swarms, and unmanned underwater vehicle (UUV) swarms. It can achieve optimized solutions within a reasonable timeframe and is applicable to various mission scenarios, thereby improving collaborative mission decision-making capabilities and efficiency.

[0005] To achieve the above objectives, this invention provides a method for coordinating and allocating inter-swarm task resources for unmanned swarm command and control, comprising:

[0006] (1) Access and parse the formation-level task plan, automatically extract all unmanned cluster tasks and the requirements of each task, and extract the task activity sequence and resource requirements of each task activity in turn based on the task's meta-task model for each task.

[0007] (2) Based on the extracted resource requirements of the task activities and the available task resource set, the feasible unmanned clusters for each single task are automatically calculated in sequence, all tasks are traversed, and a list of candidate unmanned clusters for multiple tasks is obtained.

[0008] (3) Based on the multi-task candidate unmanned cluster list, automatic multi-task resource contention detection is realized. When there is resource contention, the resource contention is automatically coordinated and allocated among unmanned clusters based on maximizing the overall benefit of the formation task. When there is no resource contention, the unmanned cluster with the highest comprehensive evaluation value is directly selected as the task resource allocation result.

[0009] (4) Based on the automatically generated inter-group task resource coordination and allocation, the task resource allocation results are displayed in the form of graphs and tables, supporting manual adjustment of resource allocation, and finally forming a formation-level task resource allocation scheme.

[0010] In some alternative implementations, step (1) includes:

[0011] (1.1) Parse the multi-domain and group-level task information in the formation-level task plan. The content extracted from each task parsing includes: task number, task name, task domain, task type, task time, task area and task objective.

[0012] (1.2) Based on the task activity sequence description in the meta-task model corresponding to each task, extract the task activity sequence of each task to obtain the task activity name and the temporal relationship of each task. The temporal relationship can be parallel or serial.

[0013] (1.3) Extract task resource requirements based on the task activity resource requirements in the meta-task model corresponding to each task.

[0014] In some alternative implementations, step (2) includes:

[0015] (2.1) Based on task information, task activity resource requirements and available task resource set, according to the granularity of task activities, and comprehensively considering task capability and comprehensive task capability index requirements, task platform type or model and task platform type or model quantity requirements, task payload type or model and task payload type or model quantity requirements, task time constraints, task space constraints, and task activity sequence time constraints, automatically calculate all candidate unmanned clusters that meet the task requirements, and calculate the individual evaluation and comprehensive assessment values ​​of the task satisfaction and task resource cost of the candidate unmanned clusters.

[0016] (2.2) Traverse all tasks, summarize the candidate unmanned clusters corresponding to each single task, and obtain a list of candidate unmanned clusters for multiple tasks.

[0017] In some alternative implementations, step (2.1) includes:

[0018] (2.1.1) Obtain the resource requirements for the task activity;

[0019] (2.1.2) Based on the resource requirements of the mission activities, select resources from the available mission resources of the formation that meet the mission capability requirements, mission platform requirements and mission payload requirements;

[0020] (2.1.3) After traversing all activities in the task, obtain the available resource set for all activities in the task;

[0021] (2.1.4) Establish a relationship matrix between task activities and available task resources according to the time sequence of task activities;

[0022] (2.1.5) Based on the temporal relationship of task activity sequence, task time constraints and task space constraints, search for available unmanned clusters in the relationship matrix to form candidate unmanned clusters;

[0023] (2.1.6) Calculate the task satisfaction, task resource cost and comprehensive evaluation of each candidate unmanned cluster.

[0024] In some alternative implementations, step (3) includes:

[0025] (3.1) Automatically detect and label the contested resources in the multi-task candidate unmanned cluster list. The judgment criteria are: whether the same type of task platform is selected as a candidate resource by multiple tasks. If not, there is no contested resource. Otherwise, it is further determined whether the number of resources selected as candidates by multiple tasks can execute multiple tasks at the same time, and there are no problems such as insufficient quantity, time conflict, space conflict, or insufficient capability. If so, there is no contested resource. Otherwise, it is indicated that there is contested resource.

[0026] (3.2) Based on maximizing the overall benefits of the formation task, an objective function for optimal allocation is established. Only in the task set where there is competition for resources, the automatic coordination and allocation of competing resources among unmanned clusters is realized. When the number of candidate unmanned clusters to be coordinated and allocated is less than the set number, the traversal combination method is directly adopted to find the coordination and allocation scheme that maximizes the objective function; otherwise, a heuristic algorithm is adopted to find the coordination and allocation scheme that maximizes the objective function.

[0027] In some alternative implementations, step (3.1) includes:

[0028] (3.1.1) Determine whether the same type of task platform has been selected as a candidate resource by multiple tasks. If not, there is no resource contention; otherwise, mark it as a potentially contentious resource.

[0029] (3.1.2) Determine whether the available quantity of potentially contested resources is greater than the total quantity of resources required by multiple tasks. If it is greater, there are no contested resources; otherwise, mark them as potentially contested resources.

[0030] (3.1.3) Determine whether the execution times of multiple tasks that may be contested for resources overlap. If the times do not overlap, further determine whether the spatial location of the task area is reachable. If it is reachable, there is no contest for resources; otherwise, mark it as a potentially contested resource. If the times of multiple tasks overlap, further determine whether the spatial locations of the task areas overlap and whether the resource task capabilities support the simultaneous support of these multiple tasks. If they overlap and the resource task capabilities support the simultaneous support of multiple tasks, there is no contest for resources; otherwise, mark it as a potentially contested resource.

[0031] In some optional implementations, the meta-task of the meta-task model represents a task that an unmanned troop can execute independently. The meta-task contains one or more task activities, which are tasks that a single unmanned platform can execute independently. The meta-task model adopts a generalized task description mechanism, defining a meta-task model for each type of task, including basic task information, task execution parameters, task execution conditions, task resource requirements, and task planning model information.

[0032] In some optional implementations, the task resource requirements in the meta-task model support three modes: based on task capabilities and indices, based on task platform type or model and the number of task platform types or models, and based on task payload type or model and the number of task payload types or models, and can describe resource requirements in one or more modes.

[0033] In some alternative implementations, the available task resource set in step (2) is constructed according to the task platform. Each model of task resource in the task resource set includes the name, platform model, platform type, platform task capability, platform task payload type, and model and available quantity information of the platform task payload type.

[0034] In some optional implementations, there are one or more candidate unmanned clusters in step (2), supporting three cluster types: unmanned aerial vehicle clusters, unmanned surface vessel clusters, and unmanned underwater vehicle clusters; homogeneous and heterogeneous unmanned platforms are supported within the cluster; each candidate unmanned cluster contains information on resource type and quantity.

[0035] In some alternative implementation schemes, the task satisfaction in step (2) can be based on metrics such as resource capacity and expected task benefits; the task resource cost can be based on metrics such as time consumption, resource usage, and total flight distance; and the comprehensive evaluation value can be based on the weighted comprehensive value of individual evaluation results.

[0036] In some alternative implementations, the heuristic algorithm in step (3) may be based on simulated annealing, genetic algorithm, list search algorithm, evolutionary programming, ant colony algorithm, etc.

[0037] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0038] This invention addresses the potential mission conflicts arising from competition for unmanned swarm resources in multi-domain operations, as well as conflicts in mission allocation. It comprehensively considers factors such as mission activity resource requirements, mission time constraints, mission space constraints, resource quantity and mission capability constraints, and mission benefits. It provides a decision-making method for coordinating and allocating inter-swarm mission resources based on maximizing overall mission benefits. This method reduces the mission resource allocation problem to a resource contention problem, enabling efficient and universal coordination and allocation of mission resources among unmanned aerial vehicle (UAV) swarms, unmanned surface vessel (USV) swarms, and unmanned underwater vehicle (UUV) swarms in multi-domain, multi-mission scenarios, thereby improving collaborative mission decision-making capabilities and efficiency. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating a method provided in an embodiment of the present invention;

[0040] Figure 2 This is a flowchart illustrating the steps of unmanned cluster computation for task candidates provided in an embodiment of the present invention;

[0041] Figure 3 This is an example diagram of the relationship matrix in the task candidate unmanned cluster calculation step provided by an embodiment of the present invention;

[0042] Figure 4 This is a flowchart illustrating the steps for coordinating and allocating task resources among groups, as provided in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0044] This invention provides a method for coordinating and allocating task resources among unmanned swarms for command and control, enabling coordinated allocation of task resources among unmanned aerial vehicle (UAV) swarms, unmanned surface vessel (USV) swarms, and unmanned underwater vehicle (UUV) swarms in multi-domain, multi-task scenarios. Specifically, it includes four main steps: task resource requirement extraction, calculation of candidate unmanned swarms, coordination and allocation of task resources among swarms, and generation of task resource allocation schemes. The overall process is as follows: Figure 1As shown.

[0045] Taking the coordination and allocation of task resources among unmanned swarms in a formation-level task plan P as an example (this formation-level task plan includes two domain tasks, denoted as D1 and D2 respectively; where D1 contains a grouping task) D2 includes three grouping tasks. This formation can utilize 7 types of mission resources, denoted as: R1, R2, R3, R4, R5, R6, and R7. The technical solution of this invention is described in detail below:

[0046] 1) Extraction of task resource requirements

[0047] It accesses and parses the formation-level task plan P, automatically extracting all unmanned cluster tasks. The requirements are as follows: For each task, based on its respective meta-task model, extract its task activity sequence and the resource requirements of each task activity in sequence. This includes three sub-steps: task extraction, task activity sequence extraction, and task activity resource requirement extraction.

[0048] A. Task Extraction: Analyzing multi-domain and group-level tasks in formation-level task plans. The task information is as follows. The content extracted from each task includes: task number T_num, task name T_name, task domain T_Domain, task type T_Type, task time <T_ts|T_preT,T_te>, task region <T_r1,T_r2,...>, and task objective <T_t1,T_t2,...>. Where T_ts represents the start time; T_preT represents the preceding task; T_te represents the end time; T_r represents a single region (a task supports one or more regions); and T_t represents a single objective (a task supports one or more objectives).

[0049] B. Task Activity Sequence Extraction: Based on the task activity sequence description in the meta-task model corresponding to each task, extract the task activity sequence to obtain the task activity name and its temporal relationship; the temporal relationship can be parallel or serial.

[0050] In this embodiment, For example, a task includes three sequential task activities, denoted as follows: For example, a task includes three parallel task activities, denoted as follows: For example, a task includes one task activity, denoted as: For example, a task includes one task activity, denoted as:

[0051] C. Extraction of Task Activity Resource Requirements: Based on the task activity resource requirements in the corresponding meta-task model for each task (the meta-task model adopts a generalized task description mechanism, with one meta-task model defined for each type of task, including basic task information, task execution parameters, task execution conditions, task resource requirements, and task planning model information), the task resource requirements are extracted.

[0052] The task resource requirements in the meta-task model support three modes: based on task capability and index, based on task platform type or model and its quantity, and based on task payload type or model and its quantity. Resource requirements can be described using one or more modes. The capability index ranges from 0 to 1; the quantity is a number greater than 1. In this embodiment, the task... For example, its task activity resource requirements are as follows:

[0053]

[0054] C A1 C A2 C A3 This indicates resource requirements based on task capabilities; (-) indicates no requirement for the capability index; (0.9) indicates the capability index should be greater than 0.9; P A1 (2) indicates resource requirements based on task platform type; L indicates a quantity requirement of 2. A3 This indicates resource requirements based on task load type; (-) indicates no quantity requirement; NuLL indicates no requirement or limit under this mode.

[0055] 2) Unmanned cluster computing for task candidates

[0056] Based on the extracted task activity resource requirements and the available task resource set, feasible unmanned clusters for each single task are automatically calculated sequentially. After traversing all tasks, a multi-task candidate unmanned cluster list is generated. This process includes two sub-steps: calculating single-task candidate unmanned clusters and generating a multi-task candidate unmanned cluster list. The workflow is as follows: Figure 2 As shown.

[0057] A. Single-task candidate unmanned cluster calculation: Based on task information, task activity resource requirements, and available task resource sets, and according to the granularity of task activities, comprehensively considering task capabilities and their exponential requirements, task platform type or model and its quantity requirements, task payload type or model and its quantity requirements, task time constraints, task space constraints, and task activity sequence temporal constraints, the system automatically calculates all candidate unmanned clusters that meet the task requirements, and calculates their task satisfaction, task resource cost, and other individual evaluations and their comprehensive assessment values. Specifically, this includes the following steps:

[0058] a) Obtain the task activity resource requirements extracted in step 1);

[0059] b) Based on the resource requirements of the mission activities, select resources from the available mission resource set (in this embodiment, it includes 7 types of available resources: R1, R2, R3, R4, R5, R6, and R7) that meet the mission capability requirements, mission platform requirements, and mission payload requirements; based on the mission... Activity For example, based on its resource requirements Resources R1, R2, and R7 were selected to meet the requirements;

[0060] c) Repeat steps 2)Aa)-2).Ab) until all activities of the task have been traversed. In this embodiment, the task is obtained. Activity The available resources are R1, R2, and R7; Activity The available resources are R1 and R4; Activity The available resources are resources R1, R2, and R3.

[0061] d) Establish a relationship matrix between task activities and available task resources according to the time sequence of task activities. For example, its relation matrix is ​​as follows: Figure 3 As shown, "0" indicates that the resource is unavailable; "1" indicates that the resource is available.

[0062] e) Based on the temporal relationship of task activity sequences, task time constraints <T_ts|T_preT,T_te>, and task space constraints <T_r1,T_r2,...>, search for available unmanned clusters in the relationship matrix obtained in 2).Ad), forming candidate unmanned clusters. In this embodiment, the task... For example, to get the task The candidate unmanned clusters are:

[0063] First candidate unmanned cluster

[0064] Second candidate unmanned cluster

[0065] The third candidate unmanned cluster

[0066] The fourth candidate unmanned cluster

[0067] f) Calculate the individual evaluations such as task satisfaction, task resource cost, and their comprehensive evaluation value E for each candidate unmanned cluster. In this embodiment, task satisfaction is calculated based on the resource capability index; task resource cost is calculated based on the resource value multiplied by the resource quantity; the comprehensive evaluation value is obtained by weighted fusion calculation of task satisfaction and task resource cost, wherein the fusion weights of task satisfaction and task resource cost are 0.7 and 0.3, respectively.

[0068] B. Generation of Multi-Task Candidate Unmanned Cluster List: Repeat step A to traverse all tasks, summarize the calculations of single-task candidate unmanned clusters, and obtain the list of multi-task candidate unmanned clusters and their comprehensive evaluation values ​​as follows:

[0069] Candidate unmanned cluster

[0070] Candidate unmanned cluster

[0071] Candidate unmanned cluster

[0072] Candidate unmanned cluster

[0073] 3) Coordination and allocation of task resources among groups

[0074] Inter-group task resource coordination and allocation: Based on a multi-task candidate unmanned cluster list, automatic multi-task resource contention detection is achieved. When resource contention exists, the resource contention is automatically coordinated and allocated among unmanned clusters to maximize the overall efficiency of the group tasks. When no resource contention exists, the unmanned cluster with the highest comprehensive evaluation value is directly selected as the task resource allocation result. Specifically, this includes two sub-steps: resource contention detection and resource contention coordination and allocation, as follows: Figure 4 As shown.

[0075] A. Resource Contention Detection: Automatically detects and marks contentious resources in the multi-task candidate unmanned cluster list. The determination is based on whether the same type of task platform is selected as a candidate resource by multiple tasks. If not, there is no contention for resources; otherwise, it further determines whether the number of resources selected as candidates by multiple tasks is sufficient to execute multiple tasks simultaneously, without issues of insufficient quantity, time conflict, space conflict, or capability insufficiency. If so, there is no contention for resources; otherwise, a message indicating that contention exists.

[0076] a) Determine whether the same type of task platform has been selected as a candidate resource by multiple tasks. If not, there is no resource contention; otherwise, mark it as a potentially contentious resource. In this embodiment, resources R1, R4, and R7 are determined by whether the same type of task platform has been selected as a candidate resource by multiple tasks, i.e., R1, R4, and R7 may be contentious resources.

[0077] b) Determine whether the available quantity of potentially contested resources in step 3).Aa) is greater than the total resource requirements of multiple tasks. If it is greater, then there are no contested resources; otherwise, mark them as potentially contested resources. In this embodiment, the available quantity of resources R1 and R4 is less than the total resource requirements of multiple tasks; the available quantity of resource R7 is greater than the total resource requirements of multiple tasks. Therefore, R1 and R4 are still potentially contested resources.

[0078] c) Determine whether the execution times of multiple tasks potentially competing for resources overlap in step 3).Ab). If the times do not overlap, further determine whether the spatial location of the task area is reachable. If reachable, there is no resource contention; otherwise, mark it as a potentially contentious resource. If multiple tasks overlap in time, further determine whether the spatial locations of the task areas overlap and whether the resource task capability supports these multiple tasks simultaneously. If they overlap and the resource task capability supports multiple tasks simultaneously, there is no resource contention; otherwise, mark it as a potentially contentious resource. In this embodiment, the task corresponding to potentially contentious resource R4 is... and Tasks that do not overlap in time and whose spatial location is accessible; but may compete for resources R1 and While the timeframes do not overlap, resource R1 has a slow movement speed and is spatially inaccessible within the mission area. Therefore, R1 is a resource-contested area.

[0079] B. Resource Contest Coordination and Allocation: Based on maximizing the overall efficiency of the formation task, an objective function for optimal allocation is established. Only in task sets where resources are contested, automatic coordination and allocation of contested resources among unmanned clusters is achieved. When the number of candidate unmanned clusters to be coordinated and allocated is less than the set number, a traversal and combination approach is directly used to find a coordination and allocation scheme that maximizes the objective function; otherwise, a heuristic algorithm is used to find a coordination and allocation scheme that maximizes the objective function.

[0080] a) Extract the task set corresponding to the contested resource and its candidate unmanned clusters. In this embodiment, the contested resource is R1, and the extracted task set to be coordinated and allocated and its candidate unmanned clusters are as follows:

[0081] Candidate unmanned cluster

[0082] Candidate unmanned cluster

[0083] b) Establish the objective function for optimal resource allocation. In this embodiment, the comprehensive evaluation value in 2).Af) is used as the objective function.

[0084] c) In this embodiment, since the number of candidate unmanned clusters is less than the set number of 20, a traversal combination method is directly used to find a coordination allocation scheme that maximizes the objective function.

[0085] Unmanned cluster: <R1×2,R2>

[0086] Unmanned cluster: <R4>

[0087] d) For other tasks where there is no competition for resources, the unmanned cluster with the highest comprehensive evaluation value is directly selected as the task resource allocation result. Thus, in this embodiment, the inter-cluster task resource coordination and allocation scheme based on maximizing overall task benefits is as follows:

[0088] Unmanned cluster: <R1×2,R2>

[0089] Unmanned swarm: <R4×2,R5>

[0090] Unmanned cluster: <R4>

[0091] Unmanned cluster: <R6,R7>

[0092] 4) Generation of task resource allocation scheme

[0093] Based on the automatically generated inter-group task resource coordination and allocation results in step 3), the task resource allocation results are displayed in the form of graphs and tables, supporting manual adjustment of resource allocation, and finally forming a formation-level task resource allocation scheme.

[0094] This invention addresses the conflict issues of resource contention and task allocation among unmanned swarms in multi-domain operations. It provides a decision-making method for coordinating and allocating task resources among swarms based on maximizing overall task benefits. This method comprehensively considers factors such as task activity resource requirements, task time constraints, task space constraints, resource quantity and task capability constraints, and task benefits. It reduces the task resource allocation problem to a resource contention problem, enabling efficient and universal coordination and allocation of task resources among unmanned aerial vehicle swarms, unmanned surface vessel swarms, and unmanned underwater vehicle swarms in multi-domain and multi-task scenarios, thereby improving collaborative task decision-making capabilities and efficiency.

[0095] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0096] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for coordinating and allocating inter-swarm task resources for command and control of unmanned swarms, characterized in that, include: (1) Access and parse the formation-level task plan, automatically extract all unmanned cluster tasks and the requirements of each task, and extract the task activity sequence and resource requirements of each task activity in turn based on the task's meta-task model for each task. (2) Based on the extracted resource requirements of the task activities and the available task resource set, the feasible unmanned clusters for each single task are automatically calculated in sequence, all tasks are traversed, and a list of candidate unmanned clusters for multiple tasks is obtained. (3) Based on the multi-task candidate unmanned cluster list, automatic multi-task resource contention detection is realized. When there is resource contention, the resource contention is automatically coordinated and allocated among unmanned clusters based on maximizing the overall benefit of the formation task. When there is no resource contention, the unmanned cluster with the highest comprehensive evaluation value is directly selected as the task resource allocation result. (4) Based on the automatically generated inter-group task resource coordination and allocation, display the task resource allocation results, support manual adjustment of resource allocation, and finally form a formation-level task resource allocation scheme; Step (2) includes: (2.1) Based on task information, task activity resource requirements and available task resource set, according to the granularity of task activities, comprehensively consider the task capability and comprehensive task capability index requirements, task platform type or model and the quantity requirements of task platform type or model, task payload type or model and the quantity requirements of task payload type or model, task time constraints, task space constraints, and task activity sequence time constraints, automatically calculate all candidate unmanned clusters that meet the task requirements, and calculate the individual evaluation and comprehensive assessment value of the task satisfaction and task resource cost of the candidate unmanned clusters. (2.2) Traverse all tasks, summarize the candidate unmanned clusters corresponding to each single task, and obtain a list of candidate unmanned clusters for multiple tasks; Step (2.1) includes: (2.1.1) Obtain the resource requirements for the task activity; (2.1.2) Based on the resource requirements of the mission activities, select resources from the available mission resources of the formation that meet the mission capability requirements, mission platform requirements and mission payload requirements; (2.1.3) After traversing all activities in the task, obtain the available resource set for all activities in the task; (2.1.4) Establish a relationship matrix between task activities and available task resources according to the time sequence of task activities; (2.1.5) Based on the temporal relationship of task activity sequence, task time constraints and task space constraints, search for available unmanned clusters in the relationship matrix to form candidate unmanned clusters; (2.1.6) Calculate the task satisfaction, task resource cost, and comprehensive evaluation value of each candidate unmanned cluster.

2. The method according to claim 1, characterized in that, Step (1) includes: (1.1) Parse the multi-domain and group-level task information in the formation-level task plan. The content extracted from each task parsing includes: task number, task name, task domain, task type, task time, task area and task objective; (1.2) Based on the task activity sequence description in the meta-task model corresponding to each task, extract the task activity sequence of each task to obtain the task activity name and the temporal relationship of each task. The temporal relationship can be parallel or serial. (1.3) Extract task resource requirements based on the task activity resource requirements in the meta-task model corresponding to each task.

3. The method according to claim 2, characterized in that, Step (3) includes: (3.1) Automatically detect and label the contested resources in the multi-task candidate unmanned cluster list. The judgment criteria are: whether the same type of task platform is selected as a candidate resource by multiple tasks. If not, there is no contested resource. Otherwise, it is further determined whether the number of resources selected as candidates by multiple tasks can execute multiple tasks at the same time, and there are no problems such as insufficient quantity, time conflict, space conflict, or insufficient capability. If so, there is no contested resource. Otherwise, it is indicated that there is contested resource. (3.2) Based on maximizing the overall benefits of the formation task, an objective function for optimal allocation is established. Only in the task set where there is competition for resources, the automatic coordination and allocation of competing resources among unmanned clusters is realized. When the number of candidate unmanned clusters to be coordinated and allocated is less than the set number, the traversal combination method is directly adopted to find the coordination and allocation scheme that maximizes the objective function; otherwise, a heuristic algorithm is adopted to find the coordination and allocation scheme that maximizes the objective function.

4. The method according to claim 3, characterized in that, Step (3.1) includes: (3.1.1) Determine whether the same type of task platform has been selected as a candidate resource by multiple tasks. If not, there is no resource contention; otherwise, mark it as a potentially contentious resource. (3.1.2) Determine whether the available quantity of potentially contested resources is greater than the total quantity of resources required by multiple tasks. If it is greater, then there is no contested resource; otherwise, mark it as a potentially contested resource. (3.1.3) Determine whether the execution times of multiple tasks that may be contested for resources overlap. If the times do not overlap, further determine whether the spatial location of the task area is reachable. If it is reachable, there is no contest for resources; otherwise, mark it as a potentially contested resource. If the times of multiple tasks overlap, further determine whether the spatial locations of the task areas overlap and whether the resource task capabilities support the simultaneous support of these multiple tasks. If they overlap and the resource task capabilities support the simultaneous support of multiple tasks, there is no contest for resources; otherwise, mark it as a potentially contested resource.

5. The method according to claim 1, characterized in that, The meta-task in the meta-task model represents a task that an unmanned group can execute independently. A meta-task contains one or more task activities, which are tasks that a single unmanned platform can execute independently. The meta-task model adopts a generalized task description mechanism. One type of task defines one meta-task model, including basic task information, task execution parameters, task execution conditions, task resource requirements, and task planning model information. The task resource requirements in the meta-task model support three modes: based on task capabilities and indices, based on task platform type or model and the number of task platform types or models, and based on task payload type or model and the number of task payload types or models. Resource requirements can be described in one or more modes.

6. The method according to claim 2, characterized in that, The available task resource set in step (2) is constructed according to the task platform. Each model of task resource in the task resource set includes the name, platform model, platform type, platform task capability, platform task payload type, and model and available quantity of platform task payload type. There are one or more candidate unmanned clusters, supporting three cluster types: drone swarms, unmanned surface vessel swarms, and unmanned underwater vehicle swarms; homogeneous and heterogeneous unmanned platforms are supported within the cluster; each candidate unmanned cluster contains information on resource type and quantity.

7. The method according to claim 6, characterized in that, In step (2), the task satisfaction can be measured based on resource capacity and expected task benefits; the task resource cost can be measured based on time consumption, resource usage and total flight distance; and the comprehensive evaluation value can be based on the weighted comprehensive value of individual evaluation results.

8. The method according to claim 3, characterized in that, The heuristic algorithm in step (3) can be based on simulated annealing, genetic algorithm, list search algorithm, evolutionary programming and ant colony algorithm.

Images