航空学报 > 2022, Vol. 43 Issue (S1): 726910-726910   doi: 10.7527/S1000-6893.2022.26910

基于决策树搜索的空间飞行器集群对抗目标分配方法

邹子缘, 陈琪锋   

  1. 1. 中南大学 航空航天学院,长沙 410083
  • 收稿日期:2022-01-07 修回日期:2022-02-24 发布日期:2022-03-04
  • 通讯作者: 陈琪锋,E-mail:chenqifeng@csu.edu.cn E-mail:chenqifeng@csu.edu.cn

Decision tree-based target assignment for confrontation of multiple space vehicles

ZOU Ziyuan, CHEN Qifeng   

  1. 1. School of Aeronautics and Astronautics, Central South University, Changsha 410083, China
  • Received:2022-01-07 Revised:2022-02-24 Published:2022-03-04

摘要: 本文针对空间飞行器集群对抗中的多对多目标分配问题,基于可达集理论计算空间飞行器集群对目标的拦截概率,建立了综合考虑可达性、燃料消耗和拦截概率的多级指标优化模型。采用决策树搜索求解该目标分配模型,为避免对相同飞行器/目标分配组合拦截概率的重复计算,提出了基于深度优先搜索生成目标分配指导表的方法,利用该指导表构建集群对抗目标分配的决策树,大大缩短了计算时间;之后基于多级指标对决策树遍历搜索,给出综合收益最优的目标分配方案。对空间飞行器集群对抗目标分配的仿真计算验证了该方法的有效性。

关键词: 决策树, 目标分配, 深度优先搜索, 可达域, 多目标优化

Abstract: To solve the problem of target assignment in confrontation of multiple space vehicles, the probability of target interception by the spacecraft cluster is calculated based on the reachable set theory. A multi-layer index optimization model considering reachability, fuel consumption and interception probability is established. Decision tree search is used to solve the target assignment model. To avoid repeated calculation of interception probability of the same interceptor-target combination, we propose a new method which use the depth-first search method to generate the target assignment guidance table, and the target assignment decision tree is generated by using the guidance table, which greatly reduces the calculation time. Then, by traversing the decision tree based on the multi-layer index, we obtain the optimal target assignment combination. The effectiveness of the proposed method is verified by simulation examples.

Key words: decision tree, target assignment, depth-first search, reachable domain, multi-objective optimization

中图分类号: