基于智能对抗进化的联合火力打击任务规划方法
|
摘要
针对常规联合火力打击任务规划方法很少涉及敌我对抗,导致评估环境发生变化的问题,提出一种基于敌我对抗进化的智能对抗进化算法。该算法以遗传算法为基础,将模拟生物竞争机制引入敌我双种群,互为评估条件实施对抗进化。依据敌我战场态势图构建观察-判断-决策-打击(OODA)超网络,计算OODA循环效率、确定敌我打击排序,通过多代对抗进化获得能够适应战场动态变化的任务规划最优个体。仿真结果表明:多代进化后的最优个体相比于标准优化结果,战场动态适应性更强,联合火力打击胜率更高,应对突发情况的响应机制更完善,能够有效地解决联合火力打击任务规划的评估优化问题。
In view of the fact that the conventional joint fire attack mission planning method rarely involves an issue of friend-foe confrontation leading to the change in evaluation environment,a smart confrontation evolution algorithm based on the evolution of friend-foe confrontation is proposed. The proposed algorithm is based on genetic algorithm,in which the simulation of biological competition mechanism is introduced into the two populations of friend and foe for implementing the confrontational evolution. An observe-orient-decide-act( OODA) super-network is constructed based on the battlefield situation map,and then the OODA cycle efficiency is calculated to determine the order of friend and foe attacks. The task-planning optimal individuals can adapt to the dynamic changes of the battlefield through the confrontation evolution of multiple generations. The simulated results show that the multi-generation evolutionary optimal individual has stronger dynamic adaptability,and the joint firepower strike rate is higher. The response mechanism to respond to the emergencies is more perfect,which can effectively solve the evaluation optimization issues of joint firepower mission planning.
In view of the fact that the conventional joint fire attack mission planning method rarely involves an issue of friend-foe confrontation leading to the change in evaluation environment,a smart confrontation evolution algorithm based on the evolution of friend-foe confrontation is proposed. The proposed algorithm is based on genetic algorithm,in which the simulation of biological competition mechanism is introduced into the two populations of friend and foe for implementing the confrontational evolution. An observe-orient-decide-act( OODA) super-network is constructed based on the battlefield situation map,and then the OODA cycle efficiency is calculated to determine the order of friend and foe attacks. The task-planning optimal individuals can adapt to the dynamic changes of the battlefield through the confrontation evolution of multiple generations. The simulated results show that the multi-generation evolutionary optimal individual has stronger dynamic adaptability,and the joint firepower strike rate is higher. The response mechanism to respond to the emergencies is more perfect,which can effectively solve the evaluation optimization issues of joint firepower mission planning.
引文
[1]贺扬清,沈治河.反舰导弹集群协同作战任务规划研究[J].舰船电子工程,2015,35(7):1-3.HE Y Q,SHEN Z H. Anti-ship missiles clustered together mission planning[J]. Ship Electronic Engineering,2015,35(7):1-3.(in Chinese)
[2]朱建明,宋彪,黄启发.基于系统动力学的网络安全攻防演化博弈模型[J].通信学报,2014,35(1):54-61.ZHU J M,SONG B,HUANG Q F. Evolution game model of offense-defense for network security based on system dynamics[J].Journal on Communications,2014,35(1):54-61.(in Chinese)
[3]张明智,马力.体系对抗OODA循环鲁棒性建模及仿真分析[J].系统仿真学报,2017,29(9):1968-1975.ZHANG M Z,MA L. System-of-systems combat OODA loop robustness modeling and experiment[J]. Journal of System Simulation,2017,29(9):1968-1975.(in Chinese)
[4]李子杰,刘湘伟.基于进化算法的多无人机协同航路规划[J].火力与指挥控制,2015,40(2):85-89.LI Z J,LIU X W. Cooperative path planning of multi-UAV based on evolutionary algorithm[J]. Fire Control&Command Control,2015,40(2):85-89.(in Chinese)
[5]张明智,马力,季明.网络化体系对抗OODA指挥循环时测建模及实验[J].指挥与控制学报,2015,1(1):50-55.ZHANG M Z,MA L,JI M. Networked system-of-systems combat OODA command circulation time measuring modeling and experiment[J]. Journal of Command and Control,2015,1(1):50-55.(in Chinese)
[6]张鑫,许峰.基于质点模型的多智能体遗传算法[J].软件导刊,2018,17(1):81-84.ZHANG X,XU F. Multi-agent genetic algorithm based on particle model[J]. Software Guide,2018,17(1):81-84.(in Chinese)
[7]李亚雄,刘新学,武健.基于改进遗传算法的多弹型混合火力分配优化模型[J].指挥控制与仿真,2017,39(4):50-54.LI Y X,LIU X X,WU J. An optimal model of multi missile hybrid firepower distribution based on improved genetic algorithm[J]. Command Control&Simulation,2017,39(4):50-54.(in Chinese)
[8]聂文亮,蔡黎,邱刚.带密度加权的自适应遗传算法[J].计算机系统应用,2018,27(1):137-142.NIE W L,CAI L,QIU G. Adaptive genetic algorithm with density weighted[J]. Computer Systems&Applications,2018,27(1):137-142.(in Chinese)
[9] HAN J K,KIM J H. Quantum-inspired evolutionary algorithms with a new termination criterion[J]. IEEE Transactions on Evolutionary Computation,2004,8(2):156-169.
[10] SPRONCK P,PONSEN M,SPRINKHUIZEN-KUYPER I,et al.Adaptive game Al with dynamic scripting[J]. Machine Learning,2006,63(3):217-248.
[11]李婵,徐龙顺,张文德.网络信息资源著作权风险演化博弈分析[J].理论研究,2018,36(1):32-37.LI C,XU L S,ZHANG W D. Evolutionary game analysis of the copyright risk of network information resources[J]. Theory Research,2018,36(1):32-37.(in Chinese)
[12]陈侠,赵明明,徐光廷.基于模糊动态博弈的多无人机空战策略研究[J].电光与控制,2014,21(6):19-24.CHEN X,ZHAO M M,XU G T. Fuzzy dynamic game based operation strategy for multiple UAVs[J]. Electronics Optics&Control,2014,21(6):19-24.(in Chinese)
[13]曾松林,王文辉,丁大春.基于动态博弈的目标分配方法研究[J].电光与控制,2011,18(2):26-29.ZENG S L,WANG W H,DING D C. Target allocation method based on dynamic game[J]. Electronics Optics&Control,2011,18(2):26-29.(in Chinese)
[14]张立仿,张喜平.量子遗传算法优化BP神经网络的网络流量预测[J].计算机工程与科学,2016,38(1):114-119.ZHANG L F,ZHANG X P. Network traffic prediction based on BP neural networks optimized by quantum genetic algorithm[J].Computer Engineering&Science,2016,38(1):114-119.(in Chinese)
[15]简平,邹鹏.混合离散粒子群的任务调度算法及应用[J].火力与指挥控制,2014,39(5):146-149.JIAN P, ZOU P. Task scheduling algorithm and application based on hybrid discrete particle swarm optimization[J]. Fire Control&Command Control,2014,39(5):146-149.(in Chinese)
[16]贾子英,闫飞龙.系统演化的作战体系对抗结构模型[J].火力与指挥控制,2013,38(9):91-94.JIA Z Y,YAN F L. Structure model for SoS combat base on system evolution[J]. Fire Control&Command Control,2013,38(9):91-94.(in Chinese)
[17]曾家有,王国卫,钟建林.多平台舰舰导弹饱和攻击几个协同问题与模型[J].兵工学报,2014,35(2):256-261.ZENG J Y,WANG G W,ZHONG J L. Research on cooperative saturation attack problems and models of ship-to-ship missiles from multi-ship platforms[J]. Acta Armamentarii, 2014,35(2):256-261.(in Chinese)
[18] SHEFFI Y. Urban transportation network:equilibrium analysis with mathematical programming methods[M]. Englewood Cliffs,NJ,US:Prentice Hall,1985:261-347.
[2]朱建明,宋彪,黄启发.基于系统动力学的网络安全攻防演化博弈模型[J].通信学报,2014,35(1):54-61.ZHU J M,SONG B,HUANG Q F. Evolution game model of offense-defense for network security based on system dynamics[J].Journal on Communications,2014,35(1):54-61.(in Chinese)
[3]张明智,马力.体系对抗OODA循环鲁棒性建模及仿真分析[J].系统仿真学报,2017,29(9):1968-1975.ZHANG M Z,MA L. System-of-systems combat OODA loop robustness modeling and experiment[J]. Journal of System Simulation,2017,29(9):1968-1975.(in Chinese)
[4]李子杰,刘湘伟.基于进化算法的多无人机协同航路规划[J].火力与指挥控制,2015,40(2):85-89.LI Z J,LIU X W. Cooperative path planning of multi-UAV based on evolutionary algorithm[J]. Fire Control&Command Control,2015,40(2):85-89.(in Chinese)
[5]张明智,马力,季明.网络化体系对抗OODA指挥循环时测建模及实验[J].指挥与控制学报,2015,1(1):50-55.ZHANG M Z,MA L,JI M. Networked system-of-systems combat OODA command circulation time measuring modeling and experiment[J]. Journal of Command and Control,2015,1(1):50-55.(in Chinese)
[6]张鑫,许峰.基于质点模型的多智能体遗传算法[J].软件导刊,2018,17(1):81-84.ZHANG X,XU F. Multi-agent genetic algorithm based on particle model[J]. Software Guide,2018,17(1):81-84.(in Chinese)
[7]李亚雄,刘新学,武健.基于改进遗传算法的多弹型混合火力分配优化模型[J].指挥控制与仿真,2017,39(4):50-54.LI Y X,LIU X X,WU J. An optimal model of multi missile hybrid firepower distribution based on improved genetic algorithm[J]. Command Control&Simulation,2017,39(4):50-54.(in Chinese)
[8]聂文亮,蔡黎,邱刚.带密度加权的自适应遗传算法[J].计算机系统应用,2018,27(1):137-142.NIE W L,CAI L,QIU G. Adaptive genetic algorithm with density weighted[J]. Computer Systems&Applications,2018,27(1):137-142.(in Chinese)
[9] HAN J K,KIM J H. Quantum-inspired evolutionary algorithms with a new termination criterion[J]. IEEE Transactions on Evolutionary Computation,2004,8(2):156-169.
[10] SPRONCK P,PONSEN M,SPRINKHUIZEN-KUYPER I,et al.Adaptive game Al with dynamic scripting[J]. Machine Learning,2006,63(3):217-248.
[11]李婵,徐龙顺,张文德.网络信息资源著作权风险演化博弈分析[J].理论研究,2018,36(1):32-37.LI C,XU L S,ZHANG W D. Evolutionary game analysis of the copyright risk of network information resources[J]. Theory Research,2018,36(1):32-37.(in Chinese)
[12]陈侠,赵明明,徐光廷.基于模糊动态博弈的多无人机空战策略研究[J].电光与控制,2014,21(6):19-24.CHEN X,ZHAO M M,XU G T. Fuzzy dynamic game based operation strategy for multiple UAVs[J]. Electronics Optics&Control,2014,21(6):19-24.(in Chinese)
[13]曾松林,王文辉,丁大春.基于动态博弈的目标分配方法研究[J].电光与控制,2011,18(2):26-29.ZENG S L,WANG W H,DING D C. Target allocation method based on dynamic game[J]. Electronics Optics&Control,2011,18(2):26-29.(in Chinese)
[14]张立仿,张喜平.量子遗传算法优化BP神经网络的网络流量预测[J].计算机工程与科学,2016,38(1):114-119.ZHANG L F,ZHANG X P. Network traffic prediction based on BP neural networks optimized by quantum genetic algorithm[J].Computer Engineering&Science,2016,38(1):114-119.(in Chinese)
[15]简平,邹鹏.混合离散粒子群的任务调度算法及应用[J].火力与指挥控制,2014,39(5):146-149.JIAN P, ZOU P. Task scheduling algorithm and application based on hybrid discrete particle swarm optimization[J]. Fire Control&Command Control,2014,39(5):146-149.(in Chinese)
[16]贾子英,闫飞龙.系统演化的作战体系对抗结构模型[J].火力与指挥控制,2013,38(9):91-94.JIA Z Y,YAN F L. Structure model for SoS combat base on system evolution[J]. Fire Control&Command Control,2013,38(9):91-94.(in Chinese)
[17]曾家有,王国卫,钟建林.多平台舰舰导弹饱和攻击几个协同问题与模型[J].兵工学报,2014,35(2):256-261.ZENG J Y,WANG G W,ZHONG J L. Research on cooperative saturation attack problems and models of ship-to-ship missiles from multi-ship platforms[J]. Acta Armamentarii, 2014,35(2):256-261.(in Chinese)
[18] SHEFFI Y. Urban transportation network:equilibrium analysis with mathematical programming methods[M]. Englewood Cliffs,NJ,US:Prentice Hall,1985:261-347.