基于聚类与时间耦合执行序列的任务分解方法

Task Decomposition Method Based on Clustering and Time Coupling Execution Sequence

摘要:

本文针对复杂任务分析中的高度耦合任务不易分解且需重构排序的问题,提出了一种基于聚类分析与改进时间-耦合执行序列的自适应任务分解方法。在矩阵最值遴选模型和任务序列转移策略相结合的基础上,设计了基于中间任务序列的任务矩阵分割算法；并进一步采用粒度自主循环调整机制,最终实现了复杂任务的自适应解耦分析。仿真验证结果表明,该方法能够有效实现复杂任务的解耦及序列重构,在作战任务分析领域具有很好的推广应用前景。

Abstract:

In this paper, an adaptive task decomposition method based on cluster analysis and improved time-coupled execution sequences is proposed for the problem that highly coupled tasks in combat mission analysis are not easily decomposed and need to be reordered. A task matrix partitioning algorithm for intermediate task sequences is designed based on the combination of a matrix-optimal selection model and a task sequence transfer strategy. At the same time, a mechanism of autonomous cyclic adjustment of granularity is further employed to finally achieve adaptive decoupling analysis of complex tasks. The simulation validation results show that the method can effectively realize the decoupling and sequence reconstruction of complex tasks, and has good prospects for application in the field of combat mission analysis.