Design and Research of Underwater Attack-Defense Algorithm Based On MoZi Platform
-
摘要:
目的 开展基于墨子平台的水下战场多智能体协同攻防算法设计研究，探索水下无人平台作战应用。 方法 提出基于墨子平台的设计思路。针对算法处理确定性规则下的行动决策，提出基于知识图谱的设计方法，采用Neo4J图数据库构建全局知识库，并设计数据交换接口，为智能算法模块提供确定性规则的辅助决策信息输入；针对算法处理非确定性和不完全信息下的行动决策，在设定场景下论述基于MADDPG的协同占位决策算法设计，提出了状态空间、动作空间、奖励函数以及与知识图谱模块的交互设计思路。 结果 介基于上述思路，完成了算法框架总体设计。在墨子平台中实现了知识图谱和智能算法的整合，并通过墨子平台提供的可视化手段，进一步验证了确定性规则下的行为约束与处理非确定性态势信息的智能算法进行整合的有效性。 结论 仿真结果表明本文研究工作可为后续进一步开展多智能体水下攻防协同算法设计提供参考。 Abstract:Objectives Carry out research on multi-agent cooperative attack-defense algorithm design of underwater battlefield based on Mozi platform, and explore the combat application of the underwater unmanned platform. Methods The design idea based on MoZi platform is proposed. For the action decision-making under deterministic rules, the design of knowledge graph module is discussed. The Neo4J database is used to build the global knowledge database, and the data exchange interface is designed to provide the auxiliary decision information input of the deterministic rules for the intelligent algorithm module. For the action decision-making under uncertain and incomplete information, the design of cooperative space occupying decision-making algorithm based on MADDPG is discussed under the set scenario, and puts forward the design idea of state space, action space, reward function and interaction with knowledge graph module. Results Based on the above mentioned ideas, the overall design of the algorithm framework is completed. The integration of knowledge graph and intelligent algorithm is realized in Mozi platform, and the effectiveness of the integration of behavior constraints under deterministic rules and intelligent algorithm for processing uncertain situation information is further verified through the visualization provided by Mozi platform. Conclusions The research work of this paper provides a reference for the further design of multi-agent underwater attack and defense cooperative algorithm. -
Key words:
- multi-agent /
- MoZi platform /
- knowledge graph /
- MADDPG
-
图 3 基于MADDPG的协同占位决策算法流程图
Figure 3. Flow chart of collaborative occupancy decision algorithm based on MADDPG
-
[1] 谢伟, 杨萌, 龚俊斌. 水下攻防对抗体系及其未来发展[J]. 中国工程科学, 2019, 21(6): 71–79. doi: 10.15302/J-SSCAE-2019.06.014XIE W, YANG M, GONG J B. Underwater attack-defense confrontation system and its future development[J]. Strategic Study of CAE, 2019, 21(6): 71–79 (in Chinese). doi: 10.15302/J-SSCAE-2019.06.014 [2] 王雅琳, 刘都群, 杨依然. 2019年水下无人系统发展综述[J]. 无人系统技术, 2020, 3(1): 55–59. doi: 10.19942/j.issn.2096-5915.2020.01.007WANG Y L, LIU D Q, YANG Y R. Summary of the development of unmanned undersea systems in 2019[J]. Unmanned Systems Technology, 2020, 3(1): 55–59 (in Chinese). doi: 10.19942/j.issn.2096-5915.2020.01.007 [3] 胡常青, 文龙贻彬, 张亚婷, 等. 基于超短基线水声定位的USV/UUV协同导航方法[J]. 中国惯性技术学报, 2019, 27(3): 327–333.HU C Q, WEN L Y B, ZHANG Y T, et al. Cooperative navigation for USV/UUV based on ultra-short baseline positioning system[J]. Journal of Chinese Inertial Technology, 2019, 27(3): 327–333 (in Chinese). [4] 蒲勇, 袁富宇. UUV群探测系统仿真设计[J]. 计算机仿真, 2012, 29(1): 143–147.PU Y, YUAN F Y. Designed of cooperative detection simulation system of UUV group[J]. Computer Simulation, 2012, 29(1): 143–147 (in Chinese). [5] 周宏坤, 葛锡云, 邱中梁, 等. UUV集群协同探测与数据融合技术研究[J]. 舰船科学技术, 2017, 39(12): 70–75.ZHOU H K, GE X Y, QIU Z L, et al. Research on UUVs cooperative detection and data fusion[J]. Ship Science And Technology, 2017, 39(12): 70–75 (in Chinese). [6] 王圣洁, 康凤举, 韩翃. 潜艇与智能无人水下航行器协同系统控制体系及决策研究[J]. 兵工学报, 2017, 38(2): 335–344.WANG S J, KANG F J, HAN H. Research on control and decision-making of submarine and intelligent UUV cooperative system[J]. Acta Armamentarii, 2017, 38(2): 335–344 (in Chinese). [7] 王昊奋, 漆桂林, 陈华钧. 知识图谱: 方法、实践与应用[M]. 北京: 电子工业出版社, 2019: 1.WANG H F, QI G L, CHEN H J. Knowledge Graph: Method, Practice and Application[M]. Beijing: Publishing House of Electronics Industry, 2019: 1 (in Chinese). [8] 江志浩, 周卿, 石敏, 等. 作战目标知识图谱构建与应用[J]. 海军航空工程学院学报, 2020, 35(6): 471–477.JIANG Z H, ZHOU Q, SHI M, et al. Construction and application of knowledge graph for target in battlefield[J]. Journal of Naval Aeronautical and Astronautical University, 2020, 35(6): 471–477 (in Chinese). [9] 梅发国, 戴大伟, 张冀. 基于知识图谱的战场目标关系融合技术[J]. 指挥信息系统与技术, 2017, 8(5): 81–86.MEI F G, DAI D W, ZHANG J. Battlefield target relation fusion technology based on knowledge mapping[J]. Command Information System and Technology, 2017, 8(5): 81–86 (in Chinese). [10] 邢萌, 杨朝红, 毕建权. 军事领域知识图谱的构建及应用[J]. 指挥控制与仿真, 2020, 42(4): 1–7.XING M, YANG C H, BI J Q. Construction and application of domain-specific knowledge graph in military field[J]. Command Control & Simulation, 2020, 42(4): 1–7 (in Chinese). [11] LOWE R, WU Y, TAMAR A, et al. Multi-agent actor-critic for mixed cooperative-competitive environments[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: ACM, 2017: 6382−6393. [12] FOERSTER J N, FARQUHAR G, AFOURAS T, et al. Counterfactual multi-agent policy gradients[C]//Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence and Thirtieth Innovative Applications of Artificial Intelligence Conference and Eighth AAAI Symposium on Educational Advances in Artificial Intelligence. New Orleans: AAAI, 2018: 363. [13] GALINDO-SERRANO A, GIUPPONI L. Distributed Q-learning for aggregated interference control in cognitive radio networks[J]. IEEE Transactions on Vehicular Technology, 2010, 59(4): 1823–1834. doi: 10.1109/TVT.2010.2043124 -
下载:
点击查看大图
图(6)
计量
- 文章访问数: 282
- HTML全文浏览量: 21
- PDF下载量: 2
- 被引次数: 0
