面向无人集群目标分配的层次化信息传播方法

doi:10.12382/bgxb.2022.0515

摘要/Abstract

摘要：

实用化集群智能构建应明确个体/群体协同协作基础性关联,形成可扩展模型与方法。当前差异化场景下模型方法难以互通共融,亟待破解共性基础,实现行为一致性与互动性统一。面向集群目标分配的层次化信息传播方法,基于复杂网络信息传播、社团结构等理论,对决策与通信解耦,构建双层耦合系统模型基础,提出基本的UAU-FO*层次化状态控制方法。面向中观尺度动态群落信息交互控制,构建基于UAU-FO*的层次化信息传播方法,形成个体目标拣选与群体协作涌现间可扩展关联,实现目标信息的集群决策传播效应和自主聚合解聚。仿真结果表明,决策与通信解耦后,通过独立信息感知融合、层间控制、群组调控策略,可实现对目标的群组行为涌现,并验证了弱通联条件下该方法具备较好鲁棒性、持续作业稳定性和系统健壮性。

关键词: 无人集群, 实用化, 复杂网络, 动态群落, 决策传播, 目标分配

Abstract:

For the construction of a practical constructing cluster intelligence, the basic associations of individual/group cooperations should be clarified to develop the extensible models and methods. At present,the differentiation models are difficultly coalesced or integrated, leading to the intense demand of taking hold of a common basis for uniting behavior consistency and interaction. Based on the theory of complex network information transmission and community structure, a hierarchical information dissemination method for cluster targets allocation is proposed for the decoupling of decision making and communication. Therefore, from a mesoscale perspective of information interaction control in and among dynamic communities, the UAU-FO* model-based hierarchical information transmission method is proposed on the basis of the double-layer coupling system model. The proposed method can be used to construct the extensible association foundation between the individual targets selection and the emergence of group cooperation, which provides a supportability for realizing the autonomous aggregation and deaggregation oriented to targets attending bythe propagation effect of decision making.The simulated results shows that the groups can achieve self-directed organization through independent information perception fusion, inter-layer control and group control strategies after the decoupling of decision and communication. Simultaneously, it is verified that the method has good robustness and continuous stability under weak interconnection conditions.

Key words: unmanned cluster, practical application, complex network, dynamic community, decision propagation, target assignment

中图分类号:

TP319

沈宇婷, 孟新, 高跃清. 面向无人集群目标分配的层次化信息传播方法[J]. 兵工学报, 2023, 44(10): 3006-3025.

SHEN Yuting, MENG Xin, GAO Yueqing. Hierarchical Information Dissemination Method for Unmanned Cluster Targets Allocation[J]. Acta Armamentarii, 2023, 44(10): 3006-3025.

图/表 23

图1 面向局域动态认知网络的智能体动力分析

Fig.1 Dynamics analysis of agents in local dynamic cognitive network

图2 基于局域动态认知的层次化系统模型

Fig.2 Hierarchical system model based on local dynamic cognition

图3 状态变迁过程

Fig.3 State transition process

图4 目标动向发布与稳态观测示意

Fig.4 Target movement publishing and steady-state observation

图5 基于群落变迁的多智能体观测集群演化

Fig.5 Multi-agent observation cluster evolution based on community change

表1 UAU-FO*层次化模型基本观测状态

Table 1 Basic observating state of UAU-FO* hierarchical model

状态	含义	详细描述
U-F	无关注自由态	节点对任何目标的信息处于无条件接收的状态,且倾向性指向不明确。
U-O*	无关注联合观测态	仅为理论状态,实际不存在,一旦节点处于对某目标的联合观测态,就将切换至有关注的相关状态,故不做讨论。
A-F	有关注且自由态	可作为实际可能存在的一种特例,尤其在任务协商博弈的多轮次过程中,有关注的自由态将是重点讨论的状态(后续将A-F归为不稳定暂态,假定一旦有了关注,就将由F过渡为O*)。
A-O*	有关注联合观测态	有观测目标,处在观测区或向观测区运动过程中,可能存在目标切换,具有OBS和ORD两种隐式状态模式切换。

图6 层次化联合观测状态分解

Fig.6 Representation of state decomposition of hierarchical joint observations

图7 UAU-FO*层次化演化状态变迁

Fig.7 UAU-FO * hierarchical evolution state transition

图8 状态控制流程

Fig.8 State control flow

图9 动态中心节点周期状态

Fig.9 Periodic state of dynamic central node

图10 基于简化UAU-FO*的层次化状态变迁

Fig.10 Hierarchical state transition based on simplified UAU-FO*

图11 算法流程

Fig.11 Algorithm flow chart

表2 基于信息交互的收益-代价评估基本指标

Table 2 Basic index of benefit-cost evaluation based on statistics of information interactions

评估指标及指标计算
距离代价I_ST,_l: $d l ∑ l = 1 s d l$
交互传播密度I_R,_l: $1 2 (e H o p 1 + e H o p 2)$ (e_Hop1 $C U ([t g t l] H o p 1) C U ([T G T t] H o p 1)$ +e_Hop2 $C U ([t g t l] H o p 2) C U ([T G T t] H o p 2)$ )
信息持久度I_ET,_l: $1 2 (e D + e F)$ (e_D $T D, l S U M (T D)$ +e_F $T F, l S U M (T F)$ ) T_D,_l=t-t_app,_l,T_F,_l=t-t_last,_l
任务匹配收益度I_RWD,_l: $1 2 (e p r i + e p r e f)$ (e_pri $w p r i, l S U M (w p r i)$ + e_pref $w p r e f, l S U M (w p r e f)$ )

表2 基于信息交互的收益-代价评估基本指标

Table 2 Basic index of benefit-cost evaluation based on statistics of information interactions

评估指标及指标计算
距离代价I_ST,_l: $d l ∑ l = 1 s d l$
交互传播密度I_R,_l: $1 2 (e H o p 1 + e H o p 2)$ (e_Hop1 $C U ([t g t l] H o p 1) C U ([T G T t] H o p 1)$ +e_Hop2 $C U ([t g t l] H o p 2) C U ([T G T t] H o p 2)$ )
信息持久度I_ET,_l: $1 2 (e D + e F)$ (e_D $T D, l S U M (T D)$ +e_F $T F, l S U M (T F)$ ) T_D,_l=t-t_app,_l,T_F,_l=t-t_last,_l
任务匹配收益度I_RWD,_l: $1 2 (e p r i + e p r e f)$ (e_pri $w p r i, l S U M (w p r i)$ + e_pref $w p r e f, l S U M (w p r e f)$ )

图12 聚合-隔离-老化-逃逸过程(LinkCapmax=20)

Fig.12 Aggregation-isolation-aging-escape process (LinkCapmax=20)

图13 聚合-隔离-老化-逃逸过程(LinkCapmax=5)

Fig.13 Aggregation-isolation-aging-escape process (LinkCapmax=5)

图14 集群通信连接分布变化(聚集-隔离)

Fig.14 Communication connection distribution of cluster (aggregation-isolation)

图15 集群通信连接分布变化

Fig.15 Communication connection distribution of cluster

图16 目标观测的群组统计

Fig.16 Statistics of targets observation group

图17 群组形成过程统计(目标2观测群组)

Fig.17 Statistics of formation process of group (observation group of Target 2)

图18 信息交互频次统计

Fig.18 Statistics of information interaction frequency

图19 目标观测的群组统计(弱通联条件)

Fig.19 Statistics of targets observation group (weak connection condition)

图20 群组形成过程统计(目标2,弱通联条件)

Fig.20 Statistics of formation process of group (for Target 2, weak connection condition)

图21 信息交互频次统计(弱通联条件)

Fig.21 Statistics of information interaction frequency (weak connection condition)

参考文献 25

[1]	朱建文, 赵长见, 李小平, 等. 基于强化学习的集群多目标分配与智能决策方法[J]. 兵工学报, 2021, 42(9): 2040-2048.
	ZHU J W, ZHAO C J, LI X P, et al. Multi-target assignment and intelligent decision based on reinforcement learning[J]. ActaArmamentarii, 2021, 42(9): 2040-2048. (in Chinese)
[2]	霍梦真, 段海滨. 基于自适应变异的多目标鸽群优化的无人机目标搜索[J]. 控制理论与应用, 2020, 37(3): 584-591.
	HUO M Z, DUAN H B. An adaptive mutant multi-objective pigeon-inspired optimization for unmanned aerial vehicle target search problem[J]. Control Theory & Applications, 2020, 37(3): 584-591. (in Chinese)
[3]	张令, 段海滨, 雍婷, 等. 基于寒鸦配对交互行为的无人机集群编队控制[J]. 北京航空航天大学学报, 2021, 47(2): 391-397.
	ZHANG L, DUAN H B, YONG T, et al. Unmanned aerial vehicle swarm formation control based on paired interaction mechanism in jackdaws[J]. Journal of Beijing University of Aeronautics and Astronautics, 2021, 47(2): 391-397. (in Chinese)
[4]	张国锋, 周凯. 基于改进鱼群算法的无人机智能突防[J]. 控制工程, 2019, 26(5): 922-926.
	ZHANG G F, ZHOU K. Intelligent penetration for UAV based on improved artificial rish swarm algorithm (AFSA)[J]. Control Engineering of China, 2019, 26(5): 922-926. (in Chinese)
[5]	齐小刚, 李博, 范英盛, 等. 多约束下多无人机的任务规划研究综述[J]. 智能系统学报, 2020, 15(2): 204-217.
	QI X G, LI B, FAN Y S, et al. A survey of mission planning on UAVs systems based on multiple constraints[J]. CAAI transactions on intelligent systems, 2020, 15(2): 204-217. (in Chinese)
[6]	王琼, 刘美万, 任伟建, 等. 无人机航迹规划常用算法综述[J]. 吉林大学学报(信息科学版), 2019, 37(1): 58-67.
	WANG Q, LIU M W, REN W J, et al. Overview of common algorithms for UAV path planning[J]. Journal of Jilin University (Information Science Edition), 2019, 37(1): 58-67. (in Chinese)
[7]	段海滨, 张岱峰, 范彦铭, 等. 从狼群智能到无人机集群协同决策[J]. 中国科学:信息科学, 2019, 49(1): 112-118.
	DUAN H B, ZHANG D F, FAN Y M, et al. From wolf pack intelligence to UAV swarm cooperative decision-making[J]. SCIENTIA SINICA Informationis, 2019, 49(1): 112-118. (in Chinese) doi: 10.1360/N112018-00168 URL
[8]	陈杰, 辛斌. 有人/无人系统自主协同的关键科学问题[J]. 中国科学:信息科学, 2018, 48(9): 1270-1274.
	CHEN J, XIN B. Key scientific problems in the autonomous cooperation of manned-unmanned systems[J]. ScientiaSinica (Informationis), 2018, 48: 1270-1274. (in Chinese)
[9]	陈杰, 樊邦奎, 邓方, 等. 智能群系统的衍化与协同——第252期双清论坛学术综述[J]. 中国科学基金, 2021, 35(4): 604-610.
	CHEN J, FAN B K, DENG F, et al. Evolution and coordination of intelligent group systems-academic summary of the 252nd ShuangqingForum[J]. Bulletin of National Natural Science Foundation of China, 2021, 35(4): 604-610. (in Chinese)
[10]	樊邦奎, 段连飞, 赵炳爱, 等. 无人机侦察目标定位技术[M]. 北京: 国防工业出版社, 2014.
	FAN B K, DUAN L F, ZHAO B A, et al. UAVreconnaissance target location technology[M]. Beijing: NationalDefence Industry Press, 2014. (in Chinese)
[11]	ZHANG T, LI Q, ZHANG C S, et al. Current trends in the development of intelligent unmanned autonomous systems[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 68-86.
[12]	DORIGO M, THERAULAZ G, TRIANNI V. Reflections on the future of swarm robotics[J]. Science Robotics, 2020, 5(49): eabe4385. doi: 10.1126/scirobotics.abe4385 URL
[13]	SIEGENFELD A F, BAR-YAM Y. An introduction to complex systems science and its applications[J]. Omplexity, 2020, 2020: 6105872.
[14]	RAHWAN I, CEBRIAN M, OBRADOVICH N, et al. Machine behaviour[J]. Nature, 2019, 568(7753): 477-486. doi: 10.1038/s41586-019-1138-y
[15]	LI S, BATRA R, BROWN D, et al. Particle robotics based on statistical mechanics of loosely coupled components[J]. Nature, 2019, 567(7748): 361-365. doi: 10.1038/s41586-019-1022-9
[16]	杨迎辉, 李建华, 王刚, 等. 基于超网络的作战信息流转建模及特性分析[J]. 复杂系统与复杂性科学, 2016, 13(3): 8-18.
	YANG Y H, LI J H, WANG G, et al. Modeling and characteristic analyzing of operational information flowing based on super-network[J]. Complex Systems and Complexity Science, 2016, 13(3): 8-18. (in Chinese)
[17]	朱涛, 梁维泰, 黄松华, 等. 面向任务的网络信息体系建模分析方法研究[J]. 系统仿真学报, 2020, 32(4): 727-737. doi: 10.16182/j.issn1004731x.joss.18-0429
	ZHU T, LIANG W T, HUANG S H, et al. Research on modeling and analyzing method of task-oriented network information system of systems[J]. Journal of System Simulation, 2020, 32(4): 727-737. (in Chinese) doi: 10.16182/j.issn1004731x.joss.18-0429
[18]	刚建勋, 叶雄兵, 王玮. 航母编队作战体系超网络建模[J]. 舰船科学技术, 2019, 41(9): 6-11.
	GANG J X, YE X B, WANG W. Hypernetwork modeling research on carrier formation fightingSoS[J]. Ship Science and Technology, 2019, 41(9): 6-11. (in Chinese)
[19]	聂俊峰, 陈行军, 史红权. 面向任务驱动的海上编队云作战体系动态超网络模型[J]. 兵工学报, 2021, 42(11): 2513-2521. doi: 10.3969/j.issn.1000-1093.2021.11.024
	NIE J F, CHEN X J, SHI H Q. Dynamic super-network model of task-oriented naval fleet “Cloud Combat” system[J]. ActaArmamentarii, 2021, 42(11): 2513-2521. (in Chinese)
[20]	DOROGOVTSEV S N, MENDES J F F. Evolution of networks with aging of sites[J]. Physical Review E, 2000, 62(2): 1842. doi: 10.1103/PhysRevE.62.1842 URL
[21]	ZHU X, WANG R, ZHU J Y. Effect of aging on network structure[J]. Physical Review E, 2003, 68(5): 056121. doi: 10.1103/PhysRevE.68.056121 URL
[22]	NEWMAN M E J, GIRVAN M. Finding and evaluating community structure in networks[J]. Physical Review E, 2004, 69(2): 026113. doi: 10.1103/PhysRevE.69.026113 URL
[23]	PALLA G, BARABÁSI A L, VICSEK T. Quantifying social group evolution[J]. Nature, 2007, 446(7136): 664-667. doi: 10.1038/nature05670
[24]	COLIZZA V, PASTOR-SATORRAS R, VESPIGNANI A. Reaction-diffusion processes and metapopulation models in heterogeneous networks[J]. Nature Physics, 2007, 3: 276. doi: 10.1038/nphys560
[25]	PASTOR-SATORRAS R, CASTELLANO C, van MIEGHEM P, et al. Epidemic processes in complex networks[J]. Reviews of Modern Physics, 2015, 87(3): 925. doi: 10.1103/RevModPhys.87.925 URL