挺进深蓝：从单体仿生到群体智能

doi:10.12263/DZXB.20201448

摘要/Abstract

摘要：

近年来，群体智能作为一项多学科融合的新技术在各领域的研究成果斐然，例如共享出行、蜂群无人机系统、水下多智能体平台等，但与水下场景结合的群体智能技术缺乏系统的归纳，有必要对水下群体智能技术的发展现状和趋势进行讨论和分析.本文对群体智能理论进行了详尽的分析，给出了群体智能的完整概念、具体算法以及应用领域.文中指出，为解决海洋复杂环境对探测、通信等造成的一系列困难，需要将群体智能技术应用于水下场景.本文就国内外水下群体智能技术的研究现状进行了总结，对水下群体智能存在的环境复杂、通信受限、信息获取困难、系统能力不足以及能量供应受限的难点进行了评述.针对这些难点，本文对结合群体智能理论的时变环境感知技术、传感网络设计、协同导航定位技术、路径规划技术、水下编队控制以及分布式自主决策技术进行了分析，并在文末给出水下群体智能技术未来在跨域通信、多平台异构、自主作业能力革新方面的发展趋势和展望.

关键词: 群体智能, 水下机器人, 智能控制, 多智能体, 自主潜航器, 粒子群算法, 分布式系统, 水下通信

Abstract:

In recent years, swarm intelligence, as a novel multi-disciplinary technology, has made remarkable achievements in various fields, such as shared traffic, swarm unmanned aerial vehicle system, underwater multi-agent platform, and so on, but the swarm intelligence technology combined with underwater scene is lack of systematic induction. Therefore, it is necessary to discuss and analyze the development status and trend of underwater swarm intelligence technology. This review aims to study and summarize the underwater swarm intelligence technology. It makes a complete introduction to the swarm intelligence theory, and gives the complete concept, specific algorithm and application field of swarm intelligence. And swarm intelligence technology needs to be applied to underwater scenes in order to solve a series of difficulties caused by complex marine environment. This review summarizes the research status of underwater swarm intelligence technology at home and abroad, and comments on the difficulties of underwater swarm intelligence, such as complex environment, limited communication, difficult information acquisition, insufficient system capacity and limited energy supply. In addition, aiming at these difficulties, this review analyzes the time-varying environment perception technology, underwater sensor network, collaborative navigation and positioning technology, path planning technology, underwater formation control and distributed autonomous decision-making technology combined with swarm intelligence theory, and gives the future application of underwater swarm intelligence technology in cross domain communication, multi-platform heterogeneous development trend and prospect of innovation of independent operation ability.

Key words: swarm intelligence, underwater robot, intelligence control, multi-agent, autonomous underwater vehicle (AUV), particle swarm optimization, distributed system, underwater communication

中图分类号:

TN929.3

陈健瑞, 王景璟, 侯向往, 方政儒, 杜军, 任勇. 挺进深蓝：从单体仿生到群体智能[J]. 电子学报, 2021, 49(12): 2458-2467.

CHEN Jian-rui, WANG Jing-jing, HOU Xiang-wang, FANG Zheng-ru, DU Jun, REN Yong. Advance into Ocean: From Bionic Monomer to Swarm Intelligence[J]. Acta Electronica Sinica, 2021, 49(12): 2458-2467.

图/表 8

参考文献 50

1	Kennedy J. Swarm Intelligence[M]. Boston: Springer, 2006.
2	Dorigo M, Birattari M, et al. Ant colony optimization[J]. IEEE Computational Intelligence Magazine, 2006, 1(4): 28 - 39.
3	Kennedy J, Eberhart R. Particle swarm optimization[A]. Proceedings of the IEEE International Conference on Neural Networks[C]. Piscataway, NJ: IEEE, 1995. 1942 - 1948.
4	Reynolds C W. Flocks, birds and schools: a distributed behavioral model[J]. Computer Graphics, 1987, 21(1): 25 - 34.
5	Mishra A K, et al. PSO-GWO optimized fractional order PID based hybrid shunt active power filter for power quality improvements[J]. IEEE Access, 2020, 8: 74497 - 74512.
6	Gandomi A H, Kashani A R. Construction cost minimization of shallow foundation using recent swarm intelligence techniques[J]. IEEE Transactions on Industrial Informatics, 2018, 14(3): 1099 - 1106.
7	Lu J, Yang C, et al. Self-tuning PID Control Scheme with Swarm Intelligence Based on Support Vector Machine[A]. 2014 IEEE International Conference on Mechatronics and Automation[C]. Piscataway, NJ: IEEE, 2014. 1554 - 1558.
8	Jun G. On dynamic evolution of industry agglomeration based on swarm intelligence[A]. 2008 International Seminar on Future Information Technology and Management Engineering[C]. Piscataway, NJ: IEEE, 2008. 24 - 28.
9	Chan K Y, Dillon T, et al. Prediction of short-term traffic variables using intelligent swarm-based neural networks[J]. IEEE Transactions on Control Systems Technology, 2013, 21(1): 263 - 274.
10	Renfrew D, Yu X. Traffic signal control with swarm intelligence[A]. 2009 Fifth International Conference on Natural Computation[C]. Piscataway, NJ: IEEE, 2009. 79 - 83.
11	Li Q, Shangguan W, et al. Traffic flow guidance and optimization of connected vehicles based on swarm intelligence[A]. 2019 Chinese Control Conference[C]. Piscataway, NJ: IEEE, 2019. 2099 - 2104.
12	Mukherjee A, Goswami P, et al. Adaptive particle swarm optimization based energy efficient dynamic correlation behavior of secondary nodes in cognitive radio sensor networks[J]. IET Communications, 2020, 14(10): 1658 - 1665.
13	孙家泽. 群体智能算法及在三维文物虚拟拼接中的应用[D]. 西安: 西北大学, 2015.
14	Bharne P K, et al. Data clustering algorithms based on swarm intelligence[A]. 2011 3rd International Conference on Electronics Computer Technology[C]. Piscataway, NJ: IEEE, 2011. 407 - 411.
15	Jin P, Zhu Y, Hu K. A clustering algorithm for data mining based on swarm intelligence[A]. 2007 International Conference on Machine Learning and Cybernetics[C]. Piscataway, NJ: IEEE, 2007. 803 - 807.
16	Kusyk J, et al. AI based flight control for autonomous UAV Swarms[A]. 2018 International Conference on Computational Science and Computational Intelligence[C]. Piscataway, NJ: IEEE, 2018. 1155 - 1160.
17	Arnold R, Jablonski J, et al. Heterogeneous UAV multi-role swarming behaviors for search and rescue[A]. 2020 IEEE Conference on Cognitive and Computational Aspects of Situation Management[C]. Piscataway, NJ: IEEE, 2020. 122 - 128.
18	李建军. 基于群体智能的多AUV协同任务分配方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2018.
19	韩青. 基于纯角度观测信息的多机器人编队控制方法研究[D]. 西安: 西北工业大学, 2018.
20	张守旭. 半仿生机器鱼建模与编队控制研究[D]. 西安: 西北工业大学, 2017.
21	Fang Z, Wang J, et al. Stochastic optimization aided energy-efficient information collection in internet of underwater things networks[J/OL]. IEEE Internet of Things Journal, 2021, DOI:10.1109/JIOT.2021.3088279.
22	Fang Z, Wang J, et al. AoI inspired collaborative information collection for AUV assisted internet of underwater things[J/OL]. IEEE Internet of Things Journal, 2021, DOI:10.1109/JIOT.2021.3049239.
23	Zhang R, Liu J. Underwater image segmentation with maximum entropy based on particle swarm optimization (PSO)[A]. First International Multi-Symposiums on Computer and Computational Sciences[C]. Piscataway, NJ: IEEE, 2006. 360 - 636.
24	Wei M, Bian H, et al. Mutual information re-registration of sensitive region in forward-looking sonar images combined with particle swarm optimization algorithm[J]. IEEE Access, 2021, 9: 7064 - 7075.
25	Rajeshwari P M, Kavitha G, et al. Swarm intelligence based segmentation for buried object scanning SONAR images[A]. 2015 IEEE Underwater Technology[C]. Piscataway, NJ: IEEE, 2015. 1 - 4.
26	Sun Y, Hu Y, et al. The coverage optimization method for underwater sensor network based on VF-PSO algorithm[A]. 2020 Chinese Control And Decision Conference[C]. Piscataway, NJ: IEEE, 2020. 2008 - 2013.
27	Su Y, Zhou Z, et al. A joint relay selection and power allocation MAC protocol for underwater acoustic sensor network[J]. IEEE Access, 2020, 8: 65197 - 65210.
28	Zhang H, Wang S, et al. A low complexity clustering optimization algorithm for underwater sensor networks[A]. 2016 IEEE International Conference on Signal Processing, Communications and Computing[C]. Piscataway, NJ: IEEE, 2016. 1 - 6.
29	Qin C, Du J, et al. A hierarchical information acquisition system for AUV assisted Internet of underwater things[J]. IEEE Access, 2020, 8: 176089 - 176100.
30	Qi Y, Wang B, et al. Cooperative navigation for multiple autonomous underwater vehicles with time delayed measurements[A]. 2016 IEEE Chinese Guidance, Navigation and Control Conference[C]. Piscataway, NJ: IEEE, 2016. 295 - 299.
31	Liu H, et al. Navigation algorithm based on PSO-BP UKF of autonomous underwater vehicle[A]. 2019 IEEE Underwater Technology[C]. Piscataway, NJ: IEEE, 2019. 1 - 4.
32	Zhuo W, Longjie J, et al. A path panning strategy for data acquisition task using multiple autonomous underwater vehicles[A]. OCEANS 2016-Shanghai[C]. Piscataway, NJ: IEEE, 2016. 1 - 5.
33	Yan S. Research on path planning of AUV based on improved ant colony algorithm[A]. 2021 4th International Conference on Artificial Intelligence and Big Data[C]. Piscataway, NJ: IEEE, 2021. 121 - 124.
34	Zhu D, Yang S X. Bio-inspired neural network-based optimal path planning for UUVs under the effect of ocean currents[J/OL]. IEEE Transactions on Intelligent Vehicles, 2021, DOI:10.1109/TIV.2021.3082151.
35	Li Y, Zhu D. Formation tracking and transformation of AUVs based on the improved particle swarm optimization algorithm[A]. 2020 Chinese Control and Decision Conference[C]. Piscataway, NJ: IEEE, 2020. 3159 - 3162.
36	Li J, Du X. Underactuated multi-AUV robust formation control based on virtual leader[A]. 2018 IEEE International Conference on Mechatronics and Automation[C]. Piscataway, NJ: IEEE, 2018. 1568 - 1573.
37	Abichandani P, Torabi S, et al. Mixed integer nonlinear programming framework for fixed path coordination of multiple underwater vehicles under acoustic communication constraints[J]. IEEE Journal of Oceanic Engineering, 2015, 40(4): 864 - 873.
38	Miftah E, Sayouti A. Multi-agent systems and their application to control vehicle underwater[A]. 2015 15th International Conference on Intelligent Systems Design and Applications[C]. Piscataway, NJ: IEEE, 2015. 353 - 357.
39	Qiuling J, Guangwen L. Formation control and obstacle avoidance algorithm of multiple autonomous underwater vehicles(AUVs) based on potential function and behavior rules[A]. 2007 IEEE International Conference on Automation and Logistics[C]. Piscataway, NJ: IEEE, 2007. 569 - 573.
40	徐玉如, 苏玉民. 关于发展智能水下机器人技术的思考[J]. 舰船科学技术, 2008, 030(004):17 - 21.
41	Gao Y, Ye J, et al. Q-learning based on particle swarm optimization for positioning system of underwater vehicles[A]. 2009 IEEE International Conference on Intelligent Computing and Intelligent Systems[C]. Piscataway, NJ: IEEE, 2009. 68 - 71.
42	Tang X, Yu F, Chen R. Path planning of underwater vehicle based on particle swarm optimization[A]. 2010 International Conference on Intelligent Control and Information Processing[C]. Piscataway, NJ: IEEE, 2010. 123 - 126.
43	Liu L, Bian H. Underwater visual tracking method for AUV based on PSOPF[A]. 2012 Second International Conference on Instrumentation, Measurement, Computer, Communication and Control[C]. Piscataway, NJ: IEEE, 2012. 985 - 989.
44	Cao Y, Yu W, et al. An overview of recent progress in the study of distributed multi-agent coordination[J]. IEEE Transactions on Industrial Informatics, 2013, 9(1): 427 - 438.
45	Champion B T, Joordens M A. Underwater swarm robotics review[A]. 2015 10th System of Systems Engineering Conference[C]. Piscataway, NJ: IEEE, 2015. 111 - 116.
46	Mahmutoglu Y, Turk K, Tugcu E. Particle swarm optimization algorithm based decision feedback equalizer for underwater acoustic communication[A]. 2016 39th International Conference on Telecommunications and Signal Processing[C]. Piscataway, NJ: IEEE, 2016. 153 - 156.
47	Li J, et al. Application of simulated annealing particle swarm optimization in underwater acoustic positioning optimization[A]. OCEANS 2019-Marseille[C]. Piscataway, NJ: IEEE, 2019. 1 - 4.
48	Cheng E, Wu L, et al. Node selection algorithm for underwater acoustic sensor network based on particle swarm optimization[J]. IEEE Access, 2019, 7: 164429 - 164443.
49	张伟, 王乃新, 等. 水下无人潜航器集群发展现状及关键技术综述[J]. 哈尔滨工程大学学报, 2020, 41(02): 289 - 297.
50	Guan S, Wang J, et al. MagicNet: the maritime giant cellular network[J]. IEEE Communications Magazine, 2021, 59(3): 117 - 123.

模型	离散/连续	对象	仿真区域
Reynolds模型	连续	同质	无界
Couzin模型	离散	异质	无界
Vicsek模型	离散	同质	重复边界

模型	离散/连续	对象	仿真区域
Reynolds模型	连续	同质	无界
Couzin模型	离散	异质	无界
Vicsek模型	离散	同质	重复边界

通信方式	电磁波通信	水声通信	激光通信
载体	电磁波	声波	蓝绿光
传播速度	2.25×10⁸m/s	1940m/s	2.25×10⁸m/s
频率/波长	3Hz~300kHz	3Hz~97kHz	450nm~560nm
应用场景	浅水近距离通信	中远距离深海通信	对潜通信
优势	速率高，功耗低	距离远，衰减小	方向性好，速率高
局限	衰减大、距离近、天线大	多径效应，环境噪声，多普勒效应，速率低、时延大	散射现象，背景辐射，吸收效应，难以对准

通信方式	电磁波通信	水声通信	激光通信
载体	电磁波	声波	蓝绿光
传播速度	2.25×10⁸m/s	1940m/s	2.25×10⁸m/s
频率/波长	3Hz~300kHz	3Hz~97kHz	450nm~560nm
应用场景	浅水近距离通信	中远距离深海通信	对潜通信
优势	速率高，功耗低	距离远，衰减小	方向性好，速率高
局限	衰减大、距离近、天线大	多径效应，环境噪声，多普勒效应，速率低、时延大	散射现象，背景辐射，吸收效应，难以对准

难点	技术支撑	应用
水下通信	跳频通信，OFDM，MIMO通信	海洋自组织网络，集群AUV通信
信息获取	光学成像，声学成像，磁探测技术	海洋科考，鱼群探测，导航定位
智能决策	分布式控制，机器学习，神经网络	编队控制，航线规划，智能避障