﻿ 基于HS-PSO算法的水下无人集群协同探测优化部署技术
 舰船科学技术  2022, Vol. 44 Issue (24): 50-55    DOI: 10.3404/j.issn.1672-7649.2022.24.011 PDF

1. 中国电子科技集团公司第二十八研究所，江苏 南京 210007;
2. 哈尔滨工程大学 水声工程学院，黑龙江 哈尔滨 150001

Optimal deployment techniques of underwater unmanned swarm for cooperative detecting
CUI Hua-chao1,2, ZHAO Yu-lin1, YAN Xie-fei1, SHENG Xue-li2
1. The 28th Research Institute of China Electronics Technology Group Corporation, Nanjing 210007, China;
2. College of Underwater Acoustic Engineering, Harbin Engineering University, Harbin 150001, China
Abstract: Using underwater unmanned swarm to carry out cooperative mission is a new operational style in the future naval battle field. The efficiency of cluster networking mainly depends on optimal deployment techniques. In order to meet cooperative detecting requirements of underwater unmanned swarm, in this paper, the combination of harmony search (HS) and particle swarm optimization (PSO) algorithm has been used to automatically calculate the optimal parameters such as swarm member’s location. It can maximize the effective detection coverage of the task area. The comprehensive evaluation algorithm of effective detection coverage under multiple constraints is established to analyze and evaluate the optimal deployment effect. The simulation result shows that the algorithm has an relatively stable global optimization ability, and the efficiency of swarm collaborative detection optimized by HS-PSO algorithm is significantly improved, which can provide strong support for intelligent cooperative task planning of underwater unmanned swarm.
Key words: HS-PSO algorithm     underwater unmanned swarm     cooperative detecting     optimal deployment
0 引　言

1 问题描述 1.1 应用场景

1.2 约束条件

1）协同探测任务要求

2）气象海洋环境

3）无人平台工作参数

1.3 评估准则

 ${P_d} = 1 - \prod\limits_{i = 1}^{m \cdot n} {(1 - {p_i})} 。$ (1)

 $S=a\cdot {u}_{1}+{b\cdot u}_{2}。$ (2)
 图 1 综合评价指标结构示意图 Fig. 1 Structure diagram of comprehensive evaluation index
2 算法设计

1）建立和声记忆库

 ${\boldsymbol{HM}} = \left[ {\begin{array}{*{20}{c}} {{x_1}}&{f({x_1})}&\vline & {x_1^p}&{f(x_1^p)} \\ {{x_2}}&{f({x_2})}&\vline & {x_2^p}&{f(x_2^p)} \\ \vdots & \vdots &\vline & \vdots & \vdots \\ {{x_{HMS}}}&{f({x_{HMS}})}&\vline & {x_{HMS}^p}&{f(x_{HMS}^p)} \end{array}} \right]。$ (3)

2）产生新的和声向量

 $\begin{split} &{x'_i} = \left\{ {\begin{array}{*{20}{l}} {{{x'}_i} \in \{ x_1^{},x_2^{}, \cdots ,x_{HMS}^{}\} }，&{rand < HMCR}, \\ {{{x'}_i} \in {X_i}}，&{{\rm{else}}} , \end{array}}\right.\\ &i = 1,2, \cdots ,HMS。\end{split}$ (4)

 $\begin{split}& {x'_i} = \left\{ {\begin{array}{*{20}{l}} {{{x'}_i} + (2 \cdot rand'' - 1) \cdot BW}，&{rand' < PAR}，\\ {{{x'}_i}}，&{{\rm{else}}} ，\end{array}} \right.\\ &i = 1,2, \cdots ,HMS。\end{split}$ (5)

3）更新和声记忆库

4）利用粒子群算法对和声进行优化

$HM$ 中的和声向量进行粒子群优化，将 $HM$ 中的每个和声当作一个粒子，依据式（6）和式（8）对各粒子的运动速度及位置更新，并给出各粒子的适应度值。

 $\begin{split} {v_i}(t) =& \omega \cdot {v_i}(t - 1) + {c_1} \cdot rand\left( {} \right) \cdot \left( {{p_i} - {x_i}(t - 1)} \right) + \\ &{c_2} \cdot rand\left( {} \right) \cdot \left( {g - {x_i}(t - 1)} \right)。\end{split}$ (6)

 $\omega {\text{ = }}\left( {{\omega _{start}} - {\omega _{end}}} \right)\left( {\frac{{{t_{\max }} - t}}{{{t_{\max }}}}} \right) + {\omega _{end}}。$ (7)

 ${x_i}(t) = {x_i}(t - 1) + {v_i}(t)。$ (8)

5）更新历史最优值

6）判断粒子群算法部分是否满足停止条件

7）判断和声搜索部分是否满足停止条件

 图 2 和声-粒子群算法实现流程图 Fig. 2 HS-PSO algorithm implementation flow chart
3 仿真试验分析 3.1 单准则约束下集群阵位优化分析

 图 3 3种算法优化处理结果对比 Fig. 3 Comparison of optimization results of three algorithms

 图 4 3种算法迭代过程曲线对比 Fig. 4 Comparison of iterative process curves of the three algorithms

 图 5 点覆盖下的区域覆盖优化前后对比 Fig. 5 Comparison of regional coverage optimization underpoint coverage
3.2 多准则约束下集群阵位优化分析

4 结　语

 [1] 生雪莉, 李鹏飞, 郭龙祥, 等. 基于单平台探测概率模型的水下无人集群部署规划算法[J]. 水下无人系统学报, 2019, 27(2): 194-199. [2] HAN Y, ZHANG J, SUN D. Error Control and Adjustment Method for Underwater Wireless Sensor Network Localization[J]. Applied Acoustics, 2018, 130: 293-299. DOI:10.1016/j.apacoust.2017.08.007 [3] WANG S Q, SUN D J, ZHANG Y W. An Efficient Intra-Cluster MAC Protocol in Underwater Acoustic Sensor Ne t-works[J]. Applied Mechanics & Materials, 2014, 651–653: 1790–1797. [4] 惠俊英, 生雪莉. 水下声信道, 第二版[M]. 北京: 国防工业出版社, 1992: 37–40. [5] 崔化超, 王益乐, 俞剑. 被动拖曳声呐模拟器关键技术[J]. 指挥信息系统与技术. 2018, 9(1): 28–32. CUI Huachao, WANG Yile, YU Jian. Key Technology of Passive Towed Sonar Simulator [J]. Command Information System and Technology. 2018, 9(1): 28-32. [6] 刘伯胜, 雷家煜. 水声学原理(第二版)[M]. 哈尔滨工程大学出版社, 2009: 23–30. [7] 佟盛, 李大辉, 戴学丰. 水下无人集群优化部署算法设计与分析[J]. 舰船科学技术. 2019, 41(11): 104–107. [8] 齐智敏, 黄谦, 张海林. 智能无人集群任务规划系统架构设计[J]. 军事运筹与系统工程. 2019, 33(3): 26–30. [9] 胡桥, 刘钰, 赵振轶, 等. 水下无人集群仿生人工侧线探测技术研究进展[J]. 水下无人系统学报. 2019, 27(2): 114–122. [10] 李宁, 陈晖. 基于灰色层次分析法的作战指挥效能评估[J]. 兵器装备工程学报. 2017, 38(5): 22–26. [11] KIM J H, GEEM Z W, KIM E S. Parameter estimation of the nonlinear Muskingum model using harmony search[J]. J Am Water Resour Assoc, 2001, 37(5): 1131-1138. DOI:10.1111/j.1752-1688.2001.tb03627.x [12] GEEM Z W. Optimal cost design of water distribution networks using harmony search[J]. Eng Optimiz, 2006, 38(3): 259-280. DOI:10.1080/03052150500467430 [13] KANG S L, GEEM Z W. A new structural optimization method based on the harmony search algorithm[J]. Comput Struct, 2004, 82(9/10): 781-798. [14] 许书诚, 王琪, 刘贤敏. 基于分合粒子群算法的多无人机任务重分配[J]. 火力与指挥控制. 2012, 37(4): 188–191. [15] 施蓉花, 吴庆宪, 姜长生. 无人机协同攻击的混合粒子群算法[J]. 火力与指挥控制. 2009, 34(9): 10–13. [16] 刘科, 周继强, 郭小和. 基于改进粒子群算法的无人机路径规划研究[J]. 中北大学学报(自然科学版). 2013, 34(4): 441–447. [17] 高立群, 葛延峰, 孔芝, 等. 自适应和声粒子群搜索算法[J]. 控制与决策. 2010, 25(7): 1101–1104. [18] 孙欣欣, 王兴伟, 黄敏. 一种基于自适应和声粒子群搜索的可信QoS路由机制[J]. 系统仿真学报. 2016, 28(3): 741–748.