A dynamic task allocation method for heterogeneous unmanned cluster in weak communication environment

Abstract

The application discloses a dynamic task allocation method for a heterogeneous unmanned cluster in a weak communication environment, and comprises the following steps: 1) loading an intelligent agent and task information, and constructing an initial communication topology; 2) optimizing the initial communication topology, identifying and removing a redundant communication link, and outputting an optimized communication topology; 3) task allocation: performing task allocation based on the optimized communication topology; and 4) generating a final task allocation scheme, including a task sequence of each intelligent agent. The application uses an improved RBM-CBBA algorithm, and has significant advantages in communication efficiency, solution efficiency, system scalability and heterogeneous adaptability, and provides a task allocation solution with higher practical value for a large-scale heterogeneous unmanned cluster system.

Description

Technical Field

[0001] This invention relates to distributed collaboration technology, and more particularly to a dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment. Background Technology

[0002] In the application of unmanned swarm technology, research and development efforts focus on expanding swarm size and improving intelligent collaboration. Among the many constraints, communication limitations have been proven to be the primary factor severely affecting the functionality of unmanned swarms, especially in complex military confrontation environments where communication challenges become even more pronounced, making traditional task allocation methods inadequate for practical needs. When heterogeneous unmanned swarm systems are applied to complex and dynamic mission environments, their efficient collaborative operation depends on a reasonable task allocation mechanism.

[0003] Consensus-Based Bundle Algorithm (CBBA) is a widely used distributed task algorithm. Studies have shown that this algorithm exhibits excellent allocation efficiency and task completion quality under conditions of low-to-medium scale systems, moderately dynamic environments, and balanced heterogeneity ratios. However, as the system scale increases, the environment becomes more dynamic, and the heterogeneity ratio becomes more complex, the performance of the dynamic CBBA algorithm shows a significant downward trend, prompting us to explore its limiting factors and optimization schemes in depth. The specific problems are: (1) The high connectivity network structure used in the traditional implementation generates a large amount of redundant communication, resulting in resource waste; (2) In highly dynamic environments, heterogeneous unmanned cluster task allocation requires more frequent information exchange, further exacerbating the system communication burden.

[0004] To address the aforementioned issues and support the efficient collaborative operation of heterogeneous unmanned swarm systems in weak communication environments, a more practical task allocation solution is needed for large-scale heterogeneous unmanned swarm systems. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment, which addresses the shortcomings of the existing technology.

[0006] The technical solution adopted by this invention to solve its technical problem is: a dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment, comprising the following steps: 1) Load agent and task information, and construct the initial communication topology. ; 2) For the initial communication topology Optimize the communication topology by identifying and removing redundant communication links. ; 3) Task allocation: based on the optimized communication topology Execute task assignment; 4) Generate the final task allocation scheme, including the task sequence of each agent and the total system score.

[0007] According to the above scheme, in step 1), the agent information includes the type, capabilities, and location of each agent in the agent set of the heterogeneous unmanned cluster.

[0008] According to the above scheme, in step 1), the task information includes task type, target location, task priority, and execution constraints.

[0009] According to the above scheme, in step 2), the initial communication topology is... Optimize to meet the following constraints: Connectivity constraints: It is fully connected, meaning that for any pair of nodes... There is a connection path for each of them; 2) Communication efficiency constraint: The performance loss of task allocation in the optimized network shall not exceed a preset threshold.

[0010] According to the above scheme, in step 2), a preset threshold is set. The calculation is as follows:

[0011] in, Represents network The task allocation performance of the CBBA algorithm is evaluated.

[0012] According to the above scheme, in step 2), redundant communication links are identified and removed, and the optimized communication topology is output. The details are as follows: Step 2.1): Edge weight calculation; assign quantized weights to each communication link to avoid removing critical links; Step 2.2) Perform triangular phantom detection to identify redundant structures in the network; Iterate through all agent triples and select the topological structures that form triangles; Step 2.3) Random pruning decision; For each detected triangle, a random number is generated based on the baseline probability to determine whether to prune; if pruning is required, the edge with the lowest weight in the triangle is removed. Baseline probability The probability of triangle pruning is controlled and set to 0.6; Step 2.4) Connectivity verification; The network is fully connected after pruning by breadth-first search (BFS), and the pruning operation is only confirmed if the connectivity is satisfied. Step 2.5) Output the optimized simplified communication topology ; While maintaining connectivity, minimize redundant links to reduce communication overhead and interference in weak communication environments.

[0013] According to the above scheme, in step 2.1), a given agent... and The weight of the edge between them is calculated using the following formula:

[0014] in, Task relevance is used to measure the degree of association between two agents in the task space. The importance of a communication link is indicated by its role in the overall flow of information. As a balancing factor, it controls the relative importance of the two factors; Task relevance The computation is based on the relationship between the agent's location and the task distribution:

[0015] in, and Representing intelligent agents respectively and The location.

[0016] According to the above scheme, in step 3), the optimized communication topology is used. The standard CBBA algorithm is used for task allocation, which includes two sub-stages: task auction and conflict resolution.

[0017] The beneficial effects of this invention are: 1. The RBM algorithm proposed in this invention cleverly introduces a communication topology optimization mechanism to accurately identify and remove redundant communication links, thereby significantly reducing communication overhead while ensuring system connectivity and allocation quality.

[0018] 2. The RBM optimization strategy of this invention provides an effective communication optimization mechanism for the CBBA algorithm. Compared with the traditional CBBA algorithm, the improved RBM-CBBA algorithm shows significant advantages in communication efficiency, solution efficiency, system scalability and heterogeneous adaptability, providing a more practical task allocation solution for large-scale heterogeneous unmanned cluster systems. Attached Figure Description

[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0021] like Figure 1 As shown, a dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment includes the following steps: 1) Load agent and task information, and construct the initial communication topology. ; Agent information includes the type, capabilities, and location of each agent in the set of agents of a heterogeneous unmanned swarm.

[0022] Task information includes task type, target location, task priority, and execution constraints; Constructing the initial communication topology To construct the initial communication topology based on the initial communication capabilities of the cluster .

[0023] 2) For the initial communication topology Optimize the communication topology by identifying and removing redundant communication links. ; The communication structure of a multi-agent system is represented as an undirected graph. , where the node set Represents intelligent agents, edge sets This represents the communication links between agents. Given an initial network... (Typically a fully connected topology), the optimization objective is to find the subgraph. ; For the initial communication topology Optimize to meet the following constraints: Connectivity constraints: It is fully connected, meaning that for any pair of nodes... There is a connection path for each of them; Communication efficiency constraint: The performance loss of task allocation in the optimized network shall not exceed a preset threshold.

[0024] Preset threshold The calculation is as follows:

[0025] in, Represents network The task allocation performance of the CBBA algorithm is evaluated.

[0026] Identify and remove redundant communication links, and output the optimized communication topology. The details are as follows: Step 2.1): Edge weight calculation; assign quantized weights to each communication link to avoid removing critical links; Edge weight calculation comprehensively considers task relevance (based on the distance decay relationship between agent position and task distribution; heterogeneous agents may affect relevance calculation due to differences in task capabilities) and communication importance (the role of links in global information flow). Given an intelligent agent and The weight of the edge between them is calculated using the following formula:

[0027] in, Task relevance is used to measure the degree of association between two agents in the task space. The importance of a communication link is indicated by its role in the overall flow of information. As a balancing factor, it controls the relative importance of the two factors; Task relevance The computation is based on the relationship between the agent's location and the task distribution:

[0028] in, and Representing intelligent agents respectively and The location.

[0029] Step 2.2) Perform triangular phantom detection to identify redundant structures in the network; Iterate through all agent triples and select the topological structures that form triangles; Step 2.3) Random pruning decision; For each detected triangle, a random number is generated based on the baseline probability to determine whether to prune; if pruning is required, the edge with the lowest weight in the triangle is removed. Baseline probability The probability of triangle pruning is controlled and set to 0.6; Step 2.4) Connectivity verification; The network is fully connected after pruning by breadth-first search (BFS), and the pruning operation is only confirmed if the connectivity is satisfied. Step 2.5) Output the optimized simplified communication topology ; While maintaining connectivity, minimize redundant links to reduce communication overhead and interference in weak communication environments.

[0030] 3) Task allocation: based on the optimized communication topology Execute task assignment; In step 3), the optimized communication topology is used. The standard CBBA algorithm is used for task allocation, which includes two sub-stages: task auction and conflict resolution.

[0031] Task auction sub-stage: Each heterogeneous intelligent agent calculates the task reward based on its own capabilities (such as the reconnaissance capability of drones and the transportation capability of unmanned vehicles), and broadcasts the bidding information through the optimized communication link. Due to the simplification of the link under weak communication, information transmission conflicts and delays are significantly reduced. Conflict resolution sub-stage: Agents exchange bidding results through local communication, resolve task competition based on the winner determination rule of CBBA (such as differences in bidding priorities of heterogeneous agents for the same task), and finally form a preliminary task allocation scheme.

[0032] 4) Generate the final task allocation scheme, including the task sequence of each agent and the total system score.

[0033] When generating the final task allocation scheme, if the weak communication environment changes (such as partial link interruption) or tasks are dynamically adjusted (adding / canceling tasks), the topology optimization and task allocation steps are repeated, and the communication topology and task scheme are updated in real time. Finally, the final task allocation scheme is generated, which clarifies the task sequence, execution priority, and total system benefits (task completion rate, resource utilization, etc.) of each heterogeneous agent, and provides feedback on the optimization effect of the communication topology.

[0034] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment, characterized in that, Includes the following steps: 1) Load agent and task information, and construct the initial communication topology. Among them, agent information includes the type, capabilities, and location of each agent in the agent set of the heterogeneous unmanned swarm; 2) For the initial communication topology Optimize the communication topology by identifying and removing redundant communication links. The details are as follows: Step 2.1) Edge weight calculation; assign quantized weights to each communication link to avoid removing critical links; Step 2.2) Perform triangular phantom detection to identify redundant structures in the network; Iterate through all agent triples and select the topological structures that form triangles; Step 2.3) Random pruning decision; For each detected triangle, a random number is generated based on the baseline probability to determine whether to prune; if pruning is required, the edge with the lowest weight in the triangle is removed. Baseline probability The probability of triangle pruning is controlled and set to 0.6; Step 2.4) Connectivity verification; Breadth-first search (BFS) is used to verify whether the network is fully connected after pruning, and the pruning operation is only confirmed if the connectivity is satisfied. Step 2.5) Output the optimized simplified communication topology ; To minimize redundant links while maintaining connectivity, and reduce communication overhead and interference in weak communication environments; 3) Task allocation: based on the optimized communication topology Execute task assignment; 4) Generate the final task allocation scheme, including the task sequence of each agent.

2. The dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment according to claim 1, characterized in that, In step 1), the task information includes task type, target location, task priority, and execution constraints.

3. The dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment according to claim 1, characterized in that, In step 2), the initial communication topology is... Optimize to meet the following constraints: Connectivity constraints: It is fully connected, meaning that for any pair of nodes... There is a connection path for each of them; Communication efficiency constraint: The performance loss of task allocation in the optimized network shall not exceed a preset threshold.

4. The dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment according to claim 1, characterized in that, In step 2.1), given an agent and The weight of the edge between them is calculated using the following formula: in, Task relevance measures the degree of association between two agents in the task space. The importance of a communication link is indicated by its role in the overall flow of information. As a balancing factor, it controls the relative importance of the two factors; Task relevance The computation is based on the relationship between the agent's location and the task distribution: in, and Representing intelligent agents respectively and The location.

5. The dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment according to claim 1, characterized in that, In step 3), the optimized communication topology is used. The standard CBBA algorithm is used for task allocation, which includes two sub-stages: task auction and conflict resolution.

6. The dynamic task allocation method for heterogeneous unmanned clusters in a weak communication environment according to claim 5, characterized in that, In step 2), a preset threshold is set. The calculation is as follows: in, Represents network The task allocation performance of the CBBA algorithm is evaluated.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 6.

Images