基于合作博弈的智能集群自主聚集策略

以无人车集群系统协同监视再入体着靶过程为任务背景,开展智能集群自组织策略相关技术研究。设计无人车集群执行再入体着靶协同监视的集群行为模式;针对协同监视过程中的集群聚集行为,提出基于合作博弈的智能集群自主聚集策略。各智能体以实现群体聚集为"合作目标",以降低自身能量消耗为"竞争目标",开展博弈;基于微粒群算法规划局部路径,最终使群体系统涌现出聚集行为。仿真实验验证了设计的自主聚集策略的有效性。     

基于判别分析阈值滤波协作通信系统信道估计

在基于正交频分复用技术的放大转发协作通信系统中,针对传统信道估计算法估计精度差的问题,提出一种基于判别分析阈值滤波离散傅立叶变换信道估计算法。该算法首先通过设置改进的阈值门限初步获得信道有效抽头,然后利用马氏距离判别分析对已检测出的信道抽头修正。仿真结果表明:和传统方法相比,提出方法有效地滤除了信道估计中的噪声,改善了信道估计精度和误码率性能。     

基于噪声调相干扰寄生扩频的隐蔽信息传输

在常见的各种信息隐蔽传输方法中,宿主载体大多集中于图像、音频、视频等各类多媒体信源,而通过射频信号作为载体来实施隐蔽信息传输的技术手段十分缺乏。针对这一情况,提出了一种基于噪声调相干扰寄生扩频的隐蔽信息传输方法。在干扰机对雷达、通信等目标对象发射噪声压制干扰信号的同时,将需要传输的信息寄生调制于噪声干扰的相位分量,形成一种类扩频传输信号,并有效隐藏于大功率干扰信号之中实现传输。在接收端通过解扩处理而恢复出隐藏的调制分量,最后解调出其中所隐蔽承载的数字比特流。仿真试验结果显示了该方法的可行性与有效性,这对于电子对抗设备的综合一体化应用和隐蔽通信的实施提供了新的技术途径。     

群决策理论在多机协同空战机动决策中的应用

针对近距空战中多架战机对空中的多个敌对目标进行协同攻击的机动决策问题进行了研究。将群决策理论引入多机协同空战机动决策,首先确定了决策成员与候选方案,然后基于战场态势评估提出了采用风险决策准则的偏好排序确定方法,给出了集结偏好的序数型群决策方法。在不同空战想定条件下进行仿真,结果表明,该方法合理、可行,具有良好的应用前景。     

基于Prony算法合闸于故障保护方案

将Prony分析方法应用于断路器合闸于故障保护。利用Matlab软件对断路器合闸于输电线路故障和非故障情况下的暂态过程进行建模和仿真,对合闸后半个周波或1个周波输电线路保护安装处实测的电流数据进行采样,应用Pronv算法对其进行分析,提取出此暂态电流中包含的各次自振荡分量的各个特征量,识别出线路中是否存在故障。仿真证明：该分析方法可在合闸后半个周波或1个周波准确求出合闸操作暂态过程中各次频率的分量,从而正确识别线路的故障。     

Re-thinking the Unthinkable: Arms Control in the Twenty-First Century

ABSTRACT

Since the end of the Cold War, arms control proponents tried to make the case for deep nuclear reductions and other forms of security cooperation as necessary for strategic stability. While different versions of strategic stability analysis did sometimes produce innovative proposals, constructive negotiations, and successful ratification campaigns in the past, this analytical framework has become more of a hindrance than a help. Treating arms control as a predominantly technical way to make deterrence more stable by changing force structure characteristics, military operations, relative numbers of weapons on either side, or total number of nuclear weapons gives short shrift to political factors, including the fundamental assumptions about world politics that inform different arms control logics, the quality of political relations among leading states, and the political processes that affect negotiation, ratification, and implementation. This article compares two logics for arms control as a means to enhance strategic stability, one developed by the Cambridge community in the 1960s and one used by the Reagan administration and its successors, with current perspectives on strategic stability in which flexibility and freedom of action are preferable to predictability and arms control. It also contrasts what the Barack Obama administration has tried to achieve through strategic stability dialogues with Russia and China with how they envision security cooperation. It then presents an approach developed during the Cold War by Hedley Bull for thinking about both the technical and the political dimensions of arms control, and suggests that the logic of Cooperative Security (which shares important features with Bull's approach) is a more appropriate and productive way to think about arms control in the twenty-first century than strategic stability analysis is.     

Cost allocation of capacity investment games

Consider a manufacturer serving a set of retail stores each of which faces deterministic demands in a finite planning horizon. At the beginning of the planning horizon, the production capacity of the manufacturer is built, followed by production, outsourcing to third party manufacturers if necessary and distribution to the retail stores. Because the retail stores are usually managed by different managers who act as independent profit centers, it is desirable that the total cost is divided among the retail stores so that their incentives can be appropriately captured and thus efficient operations can be achieved. Under various conditions, we prove that there is a fair allocation of costs among the retail stores in the sense that no subset of retail stores subsidizes others, or equivalently, the resulting capacity investment game has a nonempty core, that is, the capacity investment game is a balanced game. In addition, our proof provides a mechanism to compute a fair cost allocation. © 2013 Wiley Periodicals, Inc. Naval Research Logistics 60: 512–523, 2013     

基于协同进化的多平台联合对地观测优化调度

在分析了卫星与无人机在执行观测与资源调度上的特性差异基础上,建立了多平台联合对地观测调度问题的数学模型,提出了多平台协同进化调度算法(MPCCPSA)进行求解。MPCCPSA采用分层式协同进化架构解决了不同类型观测方案统一调度生成问题。根据不同类型平台使用特性以及观测目标集合特点,采用分治-合作策略将其分解分配到各平台,顶层的交叉、变异操作保证各种群的多样性,底层的分治、合作算子保证卫星与无人机之间保持观测能力动态互补,在确保可行解的前提下加快收敛速度。仿真实验表明该方法能够有效解决空-天基多类型平台联合观测优化调度问题。     

UUV编队协同搜索静止目标的准最优方法

考虑到操作的简单性且实际执行搜索任务时搜索力不是无限可分,将连续空间的搜索问题转换为离散空间的最优搜索问题。通过划分网格,将连续的目标位置分布离散化。根据最优搜索理论,提出了单个无人水下航行器的准最优随机搜索方法,较好地逼近理论最优值。并以此为基础,分析了无人水下航行器编队的3种协同搜索方法:集中最优搜索、分散最大概率搜索和准最优搜索。最后通过实例仿真,得出了编队准最优搜索策略的有效性、优越性和可操作性。此方法将对无人水下航行器编队的战法研究具有参考借鉴意义。     

多平台雷达有源干扰资源分配算法

根据现代干扰机特点,建立"多对多"雷达有源干扰资源分配数学模型,结合分配算法的具体应用环境,提出了基于多Agent分布协同拍卖的雷达干扰资源分配算法,实例表明该方法可行。