| 摘要: |
| 无人集群具备良好的资源调配空间以及功能弹性，能在复杂多变的环境中执行任务。无人集群执行任务时，要考虑任务的异构性、关联性、需求和资源的适配性，以及集群协同的动态响应能力等因素。针对现有研究仅考虑当前需求进行任务分配导致的系统性能振荡的问题，通过构建串行协同优化无人集群任务动态分配模型，提出基于动态规划的集群协同重叠联盟形成算法，并进行剪枝约束控制计算量，降低分配方案大幅变动所带来的调度开销，提高系统累计效用和任务完成率。在多任务、多无人机和多无人车的多种场景中的仿真实验结果表明，所提方法使得无人集群形成的任务分配联盟结构的累计效用提升了14%以上，任务完成成功率提升了13%以上。 |
| 关键词: 无人机集群 串行多任务 任务分配 重叠联盟博弈 动态规划 |
| DOI：10.20079/j.issn.1001-893x.240826001 |
|
| 基金项目:国家自然科学基金资助项目（61971439，U22B2002）；通信抗干扰全国重点实验室基础科研创新基金（稳定支持）项目（IFN20230207） |
|
| A UAV Cluster Task Allocation and Coordination Scheduling Method for Serial Multi-tasks |
| YAO Changhua,GU Mingyi,WANG Lei,CHENG Tianyuan,QI Boyu |
| (1.School of Electronic and Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China;2.TianChang Research Institute,Nanjing University of Information Science and Technology,ぁ糐P3〗Chuzhou 239300,China;3.College of Communication Engineering,Army Engineering University of PLA,Nanjing 210007,China) |
| Abstract: |
| Unmanned clusters have good resource allocation space and functional elasticity,which can perform tasks in complex and changeable environments.When executing,factors such as the heterogeneity of tasks,the correlation between tasks,the adaptability of demand and resources,and the dynamic response capability of cluster collaboration need to be considered.The authors address the problem of system performance oscillation caused by the allocation of tasks based only on current demand in existing research.By constructing a serial collaborative optimization for dynamic task allocation of unmanned swarms,the cluster collaborative overlapping alliance formation algorithm based on dynamic programming is proposed,and pruning constraint control is used to reduce the scheduling overhead caused by the large changes in the allocation scheme,to improve the cumulative utility of the system and the task completion rate.The simulation experimental results from a variety of scenarios with multiple tasks,multiple unmanned aerial vehicles and multiple unmanned ground vehicles show that the proposed method makes the cumulative utility of the task allocation alliance structure formed by the unmanned cluster increased by more than 14%,and the task completion success rate increased by more than 13%. |
| Key words: UAV cluster serial multi-task task allocation overlapping alliance game dynamic programming |
