北京理工大学 机械与车辆学院, 北京 100081
*zhenjun.ming@bit.edu.cn
收稿:2022-03-24,
纸质出版:2023-02-10
移动端阅览
丁伟, 明振军, 王国新, 等. 基于多层次LSTM网络的多智能体攻防效能动态预测模型[J]. 兵工学报, 2023,44(1):176-192.
Wei DING, Zhenjun MING, Guoxin WANG, et al. Dynamic Prediction Model Based on Multi-level LSTM Network for Multi-agent Attack and Defense Effectiveness[J]. Acta Armamentarii, 2023, 44(1): 176-192.
丁伟, 明振军, 王国新, 等. 基于多层次LSTM网络的多智能体攻防效能动态预测模型[J]. 兵工学报, 2023,44(1):176-192. DOI: 10.12382/bgxb.2022.0192.
Wei DING, Zhenjun MING, Guoxin WANG, et al. Dynamic Prediction Model Based on Multi-level LSTM Network for Multi-agent Attack and Defense Effectiveness[J]. Acta Armamentarii, 2023, 44(1): 176-192. DOI: 10.12382/bgxb.2022.0192.
针对多智能体攻防体系存在多层次耦合、无规律涌现等特征而导致难以准确预测作战效能的问题
构建一种基于多层次长短时时间记忆(LSTM)网络的多智能体攻防效能动态预测模型。明确多智能体攻防的总体框架和作战流程
通过多主体NetLogo平台模拟红蓝智能体攻防对抗过程
以获取群体结构和作战效能在不同个体决策下的多层次演化数据。利用善于处理时序特征的LSTM网络来表征个体决策、群体结构和作战效能三层间的函数映射
并基于该映射关系进一步预测未来攻防作战效能与进程。上述建模方法已在多组仿真中证实了其可行性与有效性。实验结果表明
所建模型的作战效能预测误差仅在7%以内
对多智能体攻防的作战指挥和体系建设具有指导意义。
Considering the difficulty of accurately predicting the operational effectiveness of multi-agent attack and defense (MAAD) systems due to multi-level coupling and irregular emergence
a dynamic prediction model based on multi-level LSTM network is constructed. The overall framework and operational process of the MAAD are clarified. Then
the multi-agent NetLogo platform is used to simulate the attack and defense confrontation process of red and blue agents in order to obtain multi-level evolutionary data of population structure and operational effectiveness when different decisions are made. On this basis
long short-term memory (LSTM) networks
which are effective in processing temporal features
are adopted to characterize the function mapping among the three layers of “individual decision
population structure
and operational effectiveness”
and to further predict the future attack and defense operational effectiveness and process based on the mapping relationship. The above modeling method has been proved to be feasible and effective in multiple simulations. The experimental results show that the prediction error of the model is only within 7%
which can serve as an effective guide for operational command and system development in MAAD systems.
张超省 , 王健 , 张林 , 等 . 面向复杂障碍场的多智能体系统集群避障模型 [J ] . 兵工学报 , 2021 , 42 ( 1 ): 141 - 150 .
ZHANG C S , WANG J , ZHANG L , et al. A multi-agent system flocking model with obstacle avoidance in complex obstacle field [J ] . Acta Armamentarii , 2021 , 42 ( 1 ): 141 - 150 . (in Chinese) DOI: 10.3969/j.issn.1000-1093.2021.01.016 http://doi.org/10.3969/j.issn.1000-1093.2021.01.016 A multi-agent system flocking model with obstacle avoidance in complex obstacle field is established for the obstacle avoidance of robot flock in the ground battlefield. In this model, the complex obstacle is geometrically described with polygon shape, which reduces the probability of channel narrowing or blocking caused by the least circumscribed circle(ball). Based on the artificial potential method, a mathematical model and the solving flow are presented for the interactions between agents, agent and surrounding obstacles, agent and its target. When solving the model, the rejective edge and rejective area of polygon obstacle are determined from the relationship among the outer center of agent and the geometric properties of polygon obstacle, including inner center, outer center and edge, and the capturing edge of polygon obstacle and the temporary target of agent are determined from the relationship among the outer center of agent, the edge of polygon obstacle and the target, thus overcoming the defect that the agent is usually captured by the edge of polygon obstacle. The simulated results show that the proposed model can be used to control the flocking with obstacle avoidance of multi-agent system whether for static target or dynamic target.
SATHIYARAJ R , BHARATHI A . An efficient intelligent traffic light control and deviation system for traffic congestion avoidance using multi-agent system [J ] . Transport , 2019 , 35 ( 9 ): 1 - 9 . DOI: 10.3846/transport.2019.11650 http://doi.org/10.3846/transport.2019.11650 https://journals.vilniustech.lt/index.php/Transport/article/view/11650 https://journals.vilniustech.lt/index.php/Transport/article/view/11650
GARANINA N O , SIDOROVA E A . Ontology population as algebraic information system processing based on multi-agent natural language text analysis algorithms [J ] . Programming & Computer Software , 2015 , 41 ( 3 ): 140 - 148 .
CHENG Z , LI F , ZHANG Y . Multi-agent decision support system for missile defense based on improved PSO algorithm [J ] . Journal of Systems Engineering and Electronics , 2017 , 28 ( 3 ): 514 - 525 . DOI: 10.21629/JSEE.2017.03.11 http://doi.org/10.21629/JSEE.2017.03.11 http://www.jseepub.com/EN/10.21629/JSEE.2017.03.11 http://www.jseepub.com/EN/10.21629/JSEE.2017.03.11
牛德智 , 陈长兴 , 班斐 , 等 . 基于效能评估的航空作战进程预测 [J ] . 航空学报 , 2014 , 35 ( 5 ): 1416 - 1423 . DOI: 10.7527/S1000-6893.2013.0524 http://doi.org/10.7527/S1000-6893.2013.0524 研究了航空对抗作战中空战双方的兵力变化问题。从飞机作战效能评估的角度出发,给出了在数据链支持下的作战飞机空战能力指数计算方法,并将其转化为平均战斗力水平,从数学原理上证明了二者具有等价性和一致性。建立了能够表征数据链效应的改进Lanchester方程航空作战模型,在考虑有增援的情形下对模型进行扩展。仿真分析了等效实力比和不同增援时刻对空战进程的影响,结果表明数据链具有提高作战效能和改变战局的作用。给出了有效增援时间范围,并指出了在增援作战时存在增援介入时间的"局部最优增援时刻点",为空战决策提供了参考依据。
NIU D Z , CHEN C X , BAN F , et al. Prediction on air combat process based on effectiveness evaluation [J ] . Acta Aeronautica ET Astronautica Sinica , 2014 , 35 ( 5 ): 1416 - 1423 . (in Chinese) DOI: 10.7527/S1000-6893.2013.0524 http://doi.org/10.7527/S1000-6893.2013.0524 The problem of force change for either side in air combat is studied. Starting from an effectiveness evaluation of the aircraft, an operation ability index supported by data link is given and converted to the average battle effectiveness level, which is proved to be equivalent to and consistent with the operation ability index by mathematic theory. An air combat model based on an improved Lanchester equation and reflecting function of data link is set up, which is then extended to the case of air reinforcement. Simulation analysis shows how an equivalent force ratio and different reinforcement time influence the process of air combat, which indicates that the data link can improve combat effectiveness and change combat result. Moreover, an acting reinforce time range is provided, and local optimum reinforcement time during reinforcement access time is shown, which can provide references for decision making in air combat.
PENG B , LIU S , XU L , et al. Combat process simulation and attrition forecasting based on system dynamics and multi-agent modeling [J ] . Expert Systems with Applications , 2022 , 187 : 115976 . DOI: 10.1016/j.eswa.2021.115976 http://doi.org/10.1016/j.eswa.2021.115976 https://linkinghub.elsevier.com/retrieve/pii/S0957417421013269 https://linkinghub.elsevier.com/retrieve/pii/S0957417421013269
齐智敏 , 张海林 , 伊山 , 等 . 智能无人机群体作战效能评估指标体系研究 [J ] . 舰船电子工程 , 2021 , 41 ( 9 ): 1 - 5 .
QI Z M , ZHANG H L , YI S , et al. Research on the index system of intelligent UAV group combat effectiveness evaluation [J ] . Ship Electronic Engineering , 2021 , 41 ( 9 ): 1 - 5 . (in Chinese)
刘显光 , 张晓丰 , 陈士涛 , 等 . 基于组合评估的有人/无人机协同反舰作战效能评估 [J ] . 舰船电子工程 , 2022 , 42 ( 2 ): 140 - 146 .
LIU X G , ZHANG X F , CHEN S T , et al. Evaluation of the effectiveness of man-machine/UAV cooperative anti-ship operations based on combination evaluation [J ] . Ship Electronic Engineering , 42 ( 2 ): 140 - 146 . (in Chinese)
魏武臣 , 徐硕 , 何晓夫 , 等 . 基于作战仿真推演的空-地制导弹药效能评估 [J ] . 兵工学报 , 2022 , 43 ( 增刊1 ): 107 - 114 .
WEI W C , XU S , HE X F , et al. Effectiveness evaluation of air-ground guided ammunition based on combat simulation [J ] . Acta Armamentarii , 2022 , 43 ( S1 ): 107 - 114 . (in Chinese) DOI: 10.12382/bgxb.2022.A020 http://doi.org/10.12382/bgxb.2022.A020 The combat use and effectiveness e of air-ground guided ammunition is eveluated based on combat simulation. Considering the requirements of rapid response,platform security,launch flexibility,mission efficiency and effectiveness-cost ratio of air-ground guided ammunition,an effectiveness evaluation method of air-to-ground guided ammunition based on combat simulation is proposed.A combat simulation system is built. For the combat mission of close air support,the simulation calculation is made by taking GBU-16 laser guided bomb as an example,and the effectiveness of the simulated results is analyzed by using an analytic hierarchy process method,which provides a quantitative evaluation method for the combat effectiveness evaluation of air-ground guided ammunition,and plays an important part in the argumentation and operational research of weapon system.
郭雷平 , 段文博 , 刘宇 , 等 . 基于OODA环的合成部队光电装备作战效能评估 [J ] . 兵工学报 , 2022 , 43 ( 增刊1 ): 177 - 182 .
GUO L P , DUAN W B , LIU Y , et al. Operational effectiveness evaluation of optoelectronic equipment for combined arms based on OODA Loop [J ] . Acta Armamentarii , 2022 , 43 ( S1 ): 177 - 182 . (in Chinese) DOI: 10.12382/bgxb.2022.A009 http://doi.org/10.12382/bgxb.2022.A009 The operational effectiveness of optoelectronic equipment for combined arms is evaluated based on observe-orient-decide-act (OODA) loop. The role of optoelectronic equipment based on OODA loop is analyzed,and an operational effectiveness evaluation model of optoelectronic equipment is established for the whole process of reconnaissance and attack. The index aggregation method is used to evaluate the operational effectiveness of the main optoelectronic equipment. The evaluated results show that the performance and quantity of optoelectronic equipment of observation unit have a significant impact on the combat capability of the system. When the overall combat capacity of strike unit is at a high level,the improvement of the overall combat capability by simply improving the performance of the observation unit is limited.The evaluated results of the proposed model are basically consistent with the actual situation.The results show that the combat capability of optoelectronic equipment can be objectively evaluated by using the evaluation model based on OODA loop to evaluate the combat effectiveness of optoelectronic equipment in the system-of-systems confrontation.
王瑛 , 史翔宇 , 李超 . 基于DoDAF的无人机协同作战效能评估 [J ] . 空军工程大学学报(自然科学版) , 2020 , 21 ( 6 ): 66 - 72 .
WANG Y , SHI X Y , LI C . Research on UAV cooperative combat effectiveness evaluation based on DoDAF [J ] . Journal of Air Force Engineering University (Natural Science Edition) , 2020 , 21 ( 6 ): 66 - 72 . (in Chinese)
韩月明 , 方丹 , 张红艳 , 等 . 智能无人机集群协同作战效能评估综述 [J ] . 飞航导弹 , 2020 ( 8 ): 51 - 56 .
HAN Y M , FANG D , ZHANG H Y , et al. Overview of cooperative combat effectiveness evaluation of intelligent UAV cluster [J ] . Aerodynamic Missile Journal , 2020 ( 8 ): 51 - 56 . (in Chinese)
刘国强 , 陈维义 , 程晗 , 等 . 基于BP神经网络的炮光集成武器系统作战效能评估与预测 [J ] . 海军工程大学学报 , 2019 , 31 ( 3 ): 55 - 59 .
LIU G Q , CHEN W Y , CHENG H , et al. Evaluation and prediction of combat effectiveness on weapon system with gun and laser based on BP networks [J ] . Journal of Naval University of Engineering , 2019 , 31 ( 3 ): 55 - 59 . (in Chinese)
任俊 , 胡晓峰 , 朱丰 . 基于深度学习特征迁移的装备体系效能预测 [J ] . 系统工程与电子技术 , 2017 , 39 ( 12 ): 2745 - 2749 .
REN J , HU X F , ZHU F . Effectiveness prediction of weapon equipment system-of-systems based on deep learning feature transfer [J ] . Systems Engineering and Electronics , 2017 , 39 ( 12 ): 2745 - 2749 . (in Chinese)
陈灿 , 莫雳 , 郑多 , 等 . 非对称机动能力多无人机智能协同攻防对抗 [J ] . 航空学报 , 2020 , 41 ( 12 ): 342 - 354 .
CHEN S , MO L , ZHENG D , et al. Cooperative attack-defense game of multiple UAVs with asymmetric maneuverability [J ] . Acta Aeronautica et Astronautica Sinica , 2020 , 41 ( 12 ): 342 - 354 . (in Chinese)
赵柱 , 王毅 , 樊芮锋 , 等 . 基于多主体NetLogo平台的反无人机OODA体系对抗建模 [J ] . 系统仿真学报 , 2021 , 33 ( 8 ): 1791 - 1800 . DOI: 10.16182/j.issn1004731x.joss.20-0355 http://doi.org/10.16182/j.issn1004731x.joss.20-0355 针对日益严峻的无人机或无人机蜂群威胁问题,以及反无人机作战系统设计过程中存在的体系对抗建模、总体流程设计和作战效能评估等难题,研究无人机与反无人机系统的组织架构与作战流程,提出基于多主体建模NetLogo平台的反无人机OODA(Observe-Orient-Decide-Act)体系对抗模型,利用仿真模型中主体作用的涌现性,研究OODA模型对反无人机作战的影响和规律,验证了OODA模型是影响反无人机作战效果的关键因素之一,对反无人机作战指挥及系统建设具有重要的指导意义。
ZHAO Z , WANG Y , FAN R F , et al. Modeling onanti-UAV system-of-systems combat OODA loop based on NetLogo [J ] . Journal of System Simulation , 2021 , 33 ( 8 ): 1791 - 1800 . (in Chinese)
SUDEIKAT J , STEGHOEFER J P , SEEBACH H , et al. On the combination of top-down and bottom-up methodologies for the design of coordination mechanisms in self-organising systems [J ] . Information & Software Technology , 2012 , 54 ( 6 ): 593 - 607 .
孙华飞 , 张世强 , 何孟源 , 等 . 基于图论的雷达优化部署方法 [J ] . 北京理工大学学报 , 2020 , 40 ( 9 ): 1026 - 1032 .
SUN H F , ZHANG S Q , HE M Y , et al. Radar optimal deployment based on graph theory [J ] . Transactions of Beijing Institute of Technology , 2020 , 40 ( 9 ): 1026 - 1032 . (in Chinese)
于雅洁 , 刘贤达 , 蒋启梅 , 等 . 融合LSTM-DNN的工业安全态势预测模型 [J/OL ] . 小型微型计算机系统 : http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html.
YU Y J , LIU X D , JIANG Q M , et al. Industrial security situation prediction model based on LSTM-DNN [J/OL ] . Journal of Chinese Computer Systems : http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html http://kns.cnki.net/kcms/detqil/21.1106.TP.20211213.1115.002.html. (in Chinese)
田远洋 , 徐显涛 , 彭安帮 , 等 . 训练数据量对LSTM网络学习性能影响分析 [J ] . 水文 , 2022 , 42 ( 1 ): 29 - 34 ,22.
TIAN Y Y , XU X T , PENG A B , et al. The impact of training data volume on LSTM network learning performance [J ] . Journal of China Hydrology , 2022 , 42 ( 1 ): 29 - 34 ,22. (in Chinese)
0
浏览量
263
下载量
0
CNKI被引量
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024360号