固定翼无人机群的集群和避障控制

针对传统的集群控制算法需要获取通信范围内相邻质点的位置和速度信息才能够计算控制量的问题,提出一种新的无须获得相邻无人机速度的六自由度固定翼无人机群的集群和避障控制方法。将通信范围内的无人机均视为障碍物,采用统一的计算方法获得控制量,并且证明了算法的稳定性。通过建立六自由度无人机线性化控制模型,将改进的质点集群算法应用于无人机群控制系统中,将无人机控制设计成六自由度无人机的跟踪回路和质点无人机的导引回路,并证明通过选取合适的跟踪回路控制参数,确保整个无人机集群控制是稳定的。通过六自由度无人机编队仿真验证了所提算法的有效性。     

节点实时性能自适应的集群资源分配算法

由于配置和所运行作业的不同,集群各节点的实时性能差异较大。为提高集群性能,提出节点实时性能自适应的集群资源分配算法 (node real-time performance adaptive cluster resource scheduling algorithm,NPARSA)。节点实时性能用其配置(CPU核数及速度、内存容量、磁盘容量)和实时状态参数(CPU、内存和磁盘的剩余数量及磁盘读写速度)表示。NPARSA根据作业类型自主选择节点性能评价指标的权值,实现节点实时性能对于作业类型的自适应。实时性能最优的节点分配给作业。虚拟机实验和物理集群实验表明,与Spark默认资源分配算法、没有考虑作业类型与节点匹配的算法、使用作业和节点匹配差异程度作为资源分配依据的算法相比,NPARSA能更有效地缩短作业执行时间、提高集群性能。     

蒙特卡洛剖分投影法的爆震弹破片平均比动能评估

为了评估爆震弹封装壳体破片致死、致伤半径,并削弱其平均比动能,基于蒙特卡洛剖分投影法建立破片平均比动能计算模型。通过LS-DYNA及自编程联合仿真方法对爆震弹自然破片、半预制破片比动能计算模型进行仿真求解,获得了全破片全时域的质量分布、初始速度、垂直靶分布、平均比动能阈值及安全半径等指标。结果表明:在相同装药参数下,半预制破片相比于自然破片,质量分布、初始速度阈值更低且更集中;在小于等于2.5 m范围内半预制破片平均比动能阈值比自然破片更低,大于2.5 m后半预制破片平均比动能阈值比自然破片更高。相比于自然破片,采用半预制破片可显著减小爆震弹破片的致死半径,但并不能明显增大其安全半径。     

紫外光非视距后勤应急通信系统研究

自由空间紫外光非视距通信能够在复杂气候条件下、于复杂地形区域为一定范围的集群后勤保障行动提供应急信息支持。建立了自由空间紫外光非视距通信系统大气信道基本模型,分析了系统光束孔径角和通信距离对通信效果的影响,探讨了提高系统通信能力的策略。模拟结果证明：紫外光非视距通信系统在半径1km左右范围内能够保持较好的通信效果,满足小范围集群后勤应急保障需要。该结果可为发展复杂环境下后勤保障应急通信系统提供相应借鉴。     

针对无人机集群对抗的规则与智能耦合约束训练方法

基于无人机集群智能攻防对抗构想,建立了无人机集群智能攻防对抗仿真环境。针对传统强化学习算法中难以通过奖励信号精准控制对抗过程中无人机的速度和攻击角度等问题,提出一种规则与智能耦合约束训练的多智能体深度确定性策略梯度(rule and intelligence coupling constrained multi-agent deep deterministic policy gradient, RIC-MADDPG)算法,该算法采用规则对强化学习中无人机的动作进行约束。实验结果显示,基于RIC-MADDPG方法训练的无人机集群对抗模型能使得红方无人机集群在对抗中的胜率从53%提高至79%,表明采用“智能体训练—发现问题—编写规则—再次智能体训练—再次发现问题—再次编写规则”的方式对优化智能体对抗策略是有效的。研究结果对建立无人机集群智能攻防策略训练体系、开展规则与智能相耦合的集群战法研究具有一定参考意义。     

水中微气泡上浮过程的力学影响因子研究

从受力分析的角度,对气泡运动方程进行了推导,并对半径在40~500μm之间的微小气泡上浮过程进行了仿真计算。结果表明,当半径小于500μm、气泡上浮距离大于0.1 m时,可忽略加速过程对气泡上浮速度的影响。在此基础上,对气泡运动方程进行了详细分析,结果表明,阻力系数与粘滞系数是影响气泡上浮速度的主要因素,它们的取值与上浮速度成反比,并得出了Re≤150条件下的微气泡运动通用的阻力系数表达式。     

AWGN信道下线性分组码的差错概率下界分析

提出了一种线性分组码的最大似然译码(ML-decoding)差错概率下界的计算方法。差错概率的下界优化实质上是对联合事件概率下界的优化,算法结合改进的Dawson-Sankoff界的优化准则,提出了AWGN信道下线性分组码差错冗余事件的判决准则,得到了误码率下界的计算表达式。该表达式只依赖码字的Hamming重量分布与信噪比,较之类deCaens界与类KAT界,本算法得到的下界更紧,计算量更低。针对LDPC等线性分组码的数值结果证明了算法的优越性能。     

多属性决策的战时集群装备维修保障需求优先级评估

集群装备维修保障需求优先级评估是战时集群装备维修保障决策的前提和基础.通过对集群装备维修保障需求的分析和量化,依据多属性决策理论和方法,建立了集群装备维修保障需求优先级评估的属性集和模型,把战损装备的客观信息与决策者的主观权重偏好信息有机集成,并通过示例介绍了集群装备抢修排序的求解过程.研究结果表明,该方法是可行有效的...     

线性方程组二级迭代法的收敛速度

二级迭代法由内、外迭代和内迭代次数三部分组成。给出了线性方程组二级迭代法R1-收敛因子的一个上界,这个上界由内、外迭代的R1-收敛因子和内迭代次数所决定,其主部为外迭代的R1-收敛因子。在矩阵单调性条件下,对于任何内迭代方法和任意内迭代次数,证明了外迭代的R1-收敛因子也是二级迭代法R1-收敛因子的下界。所得结果反映了内、外迭代的收敛速度以及内迭代次数对于二级迭代法收敛速度的综合影响。     

航空集群系统空对地作战效能涌现方法研究

航空集群作战系统通过自组织机制与行为调控机制的有机耦合,产生了新的作战能力,具有能力涌现特点.为评估集群系统作战效能,在机群作战效能评估理论的基础上,采用AHP层次分析法构建了无人-有人机协同作战集群效能涌现模型.结果表明,作战环境威胁越高,集群作战能力的涌现效果越明显,且随着等级的不断提高,能力的涌现不断增加,但是增加的幅度逐渐减缓.该方法的应用可为评估航空集群系统的对地攻击作战能力提供定量参考.     

Clustering in defence-related procurement: the case of a Belgian naval construction cluster

ABSTRACT

This paper examines the importance of closer co-operation in defence-related procurement. Led by the decision that Belgium will lead the procurement of 12 Mine CounterMeasures Vessels (MCMV), the case of a potential Belgian naval construction cluster is discussed. The feasibility of a potential cluster is investigated by looking at key elements of a successful naval construction cluster, the Dutch naval construction cluster, and comparing those key elements to the current Belgian situation. Forming a sustainable Belgian naval construction cluster will be difficult. Most of the procurement process of the MCMVs is set on a European tender. In the long run, due to the absence of both a launching customer and leader firms, no party is able to carry the cluster with investments and knowledge spill-overs. In order to keep public support for large defence-related investments, clustering around the MCMV integration system and proactive cluster support is recommended.     

On efficiency of multistage channel with bargaining over wholesale prices

This article considers a multistage channel with deterministic price‐sensitive demand. Two systems for pricing decisions, that is, the bargaining system and the leader‐follower system, are compared. We characterize the necessary and sufficient conditions on the power structure, under which the solution of the bargaining system Pareto dominates that of the leader‐follower system. Also, under such conditions, we give a tight upper bound of channel efficiency of the bargaining system, which converges to 100% channel efficiency as the number of stages increases to infinity. © 2016 Wiley Periodicals, Inc. Naval Research Logistics 63: 449–459, 2016     

运动矢径对AUV协同定位精度影响的仿真

针对在基于运动矢径的自主水下航行器(AUV)协同定位算法中运动矢径对定位精度的影响进行了仿真分析.利用扩展卡尔曼滤波设计了AUV协同定位算法,通过设定不同的运动矢径对该协同定位算法进行仿真,分析研究了运动矢径的变化对定位精度的影响.仿真结果表明,该定位算法受运动矢径的影响,增大运动矢径会使滤波误差波动变大,降低定位精度.     

子母弹抛撒半径对机场毁伤效果影响分析

首先阐述了机场目标在未来战争中的地位与作用 ,在对机场目标及其子目标分析的基础上 ,建立了炮道失效率的计算模型。利用蒙卡模拟的方法重点研究了子母弹抛撒半径对机场毁伤效果的影响 ,并摸索了一些规律     

Strategic deficiencies in national liberation struggles: The case of fatah in the al-Aqsa Intifada

This study reports and explains a cluster of deviations from the basic rational criteria of national liberation strategy exhibited by ‘inside’ West Bank Fatah leaders during the al-Aqsa uprising, based on an analysis of public statements of three such leaders. The leaders fail to recognize that their attempt to deter Israeli offensives by threatening to reciprocate them with attacks inside the Green Line is sabotaged by Islamists independently attacking inside the Green Line; inadequately attend to the distinct possibility that attacks within the Green Line increase Israeli opposition to desired concessions on refugees and territory; and appear to occasionally get swept-up in the sentiment that reciprocating Israeli aggression is inherently just. The study elaborates and examines the possible roles in these strategic deficiencies of leader strategic desperation; rage and indignation; and the political need to satisfy widespread popular militancy. The study's logic complements existing asymmetric conflict research and can inform research beyond the Palestinian–Israeli case.     

机动目标跟踪误差CRLB计算与分析

克拉美-罗下限(CRLB)作为参数估计所能达到的最低误差限,常被用来衡量算法估计性能的优劣程度。以转弯机动为例,研究了有源(主动)和无源(被动)两种观测条件下目标跟踪CRLB计算的一般方法,并通过矩阵理论分析了目标机动可能引起的CRLB突变的原因,为机动目标跟踪的算法性能评估提供了有效手段。最后给出了仿真算例。     

Understanding the ‘failure’ of the Séléka rebellion

This paper examines the downfall of the Séléka rebellion, which staged a coup in the Central African Republic in March 2013. The coup plunged the country into violence and chaos, and there has been an uneasy quest for peace ever since. This article explains why the leader of the rebellion lost control over his troops and was finally pushed by regional actors to leave power. The article concludes that the Séléka rebellion was already fragile and, together with the poor leadership demonstrated by the coalition leader, the rebellion was unable to hold on to power.     

BSC信道下线性分组码的差错概率下界分析

针对BSC信道,提出了一种线性分组码的最大似然译码差错概率下界的计算方法.根据最大似然译码算法原理,首先将译码差错概率转化为差错事件的联合概率,基于改进的Dawson-Sankoff界的优化准则,推导出BSC信道下线性分组码差错冗余事件的判决准则,最后得到差错概率下界的计算表达式.该下界只依赖于码字的Hamming重量...     

A primal-dual algorithm for the dynamic lotsizing problem with joint set-up costs

In this article the multi-item dynamic lotsizing problem with joint set-up costs (LPJS) is considered. A tight formulation of the problem and the dual of the linear relaxation of this formulation is presented. A procedure to solve this dual problem is developed where the solution provides a strong lower bound for the LPJS. This lower bound is used in a branch and bound procedure to find an optimal solution to the problem. The extensive computational experiments with this procedure revealed that it outperforms the other available branch and bound algorithm in the literature. With the proposed algorithm the average time requirement was about 113 seconds on a 486DX33 processor for solving a difficult class of LPJS problems that have 50 items and 24 periods. © 1995 John Wiley & Sons, Inc.