实时位置反馈的多机器人主从式编队控制

提出了一种基于反馈控制的多机器人的编队控制方法,能够使主从机器人在相对运动时保持期望的角度和距离,并且形成稳定的编队结构。然后运用数学图论将单一主从式编队控制扩展到多机器人的编队控制,能够更加简易的实现编队控制和几何形状的变换,最后通过仿真验证了该方法的有效性。     

改进型灰色聚类法对航电火控系统仿真可信度评估

综合航电火控系统仿真的可信度研究越来越成为航电综合化仿真方面的一个重要研究课题,灰色聚类法是对综合航电火控系统仿真可信度进行评估的一种有效方法,首先对灰色理论和灰色三角白化权函数聚类评估法作了阐述,然后对改进型的灰色三角白化权函数聚类模型进行了重点说明并应用此法对综合航电火控系统进行仿真可信度评估。此方法简单易行,合理性强,为综合航电火控系统仿真可信度评估提供了一种较好的标准。     

势场理论的多无人机协同路径规划方法

在势场理论的基础上对穿越多威胁区的无人机路径进行了协同规划,在规划中将每个威胁区按实际情况分为极危险区域和次危险区域,而这些威胁不仅包括己方情报所得的已知威胁,而且包括在执行攻击任务过程中新发现的突发威胁,然后利用物理学中的扩散方程建立势场,并在已建立的势场中对多无人机进行整体协同路径规划。在建立模型的同时,进行了仿真模拟,仿真结果表明了该方法的正确性和可行性。     

摧毁固定目标的空袭兵器需求量算法研究

在反空袭作战中,预先对敌空袭兵器的出动量的有效判断程度,关系到我防空兵器的兵力需求论证与战斗部署。以往这方面的研究成果中,定性多于定量,为提高准确性,加强定量研究有很大的必要。由于影响作战的因素很多,且需求量的计算分为多个方面,不可能一一作出研究。分析了敌空袭固定目标时兵力选择所依据的因素,分别用排队论与Lanchester方程理论进行了数学建模。     

关于值分布论中一个著名猜测的推广

运用Nevanlinna的亚纯函数理论方法,研究了超越亚纯函数的值分布理论,获得了如下的结论:设f为超越亚纯函数,c为f的不恒等于0的的小函数,则当n≥3时,fnf′-c有无穷多个零点;若附加条件f只有有限多个级≤2的零点,则对一切正整数n,fnf′-c都有无穷多个零点.因而对Chiang Y M的问题作出了部分回答.     

基于RS理论的电子对抗装备编配辅助决策方法

根据电子对抗装备编配的主要特点,提出利用RS(RoughSet)理论探索装备编配问题。综合考虑影响装备编配的各种因素,建立起知识表达系统,并且约简得到一些相对简单的编配规则;然后再根据实际情况,结合数学规划方法进行装备编配优化。分析结果表明,运用RS理论,能够较好地用定性与定量的方法解决电子对抗装备编配问题。     

新形势下我国对外贸易比较优势理论分析

我国入世以后,有必要重新审视对外贸易的比较优势,迎接经济全球化的冲击与挑战。本文从资源禀赋、后起者优势及制度创新三个层面对我国的对外贸易比较优势理论予以拓展,以求为我国的开放型经济建设提供一些有益的参考。     

Bending and stress analysis of polymeric composite plates reinforced with functionally graded graphene platelets based on sinusoidal shear-deformation plate theory

The bending and stress analysis of a functionally graded polymer composite plate reinforced with gra-phene platelets are studied in this paper. The governing equations are derived by using principle of virtual work for a plate which is rested on Pasternak's foundation. Sinusoidal shear deformation theory is used to describe displacementfield. Four different distribution patterns are employed in our analysis. The analytical solution is presented for a functionally graded plate to investigate the influence of important parameters. The numerical results are presented to show the deflection and stress results of the problem for four employed patterns in terms of geometric parameters such as number of layers, weight fraction and two parameters of Pasternak's foundation.     

On vibration analysis of functionally graded carbon nanotube reinforced magneto-electro-elastic plates with different electro-magnetic conditions using higher order finite element methods

This article deals with evaluating the frequency response of functionally graded carbon nanotube rein-forced magneto-electro-elastic (FG-CNTMEE) plates subjected to open and closed electro-magnetic cir-cuit conditions. In this regard finite element formulation has been derived. The plate kinematics adjudged via higher order shear deformation theory (HSDT) is considered for evaluation. The equations of motion are obtained with the help of Hamilton's principle and solved using condensation technique. It is found that the convergence and accuracy of the present FE formulation is very good to address the vibration problem of FG-CNTMEE plate. For the first time, frequency response analysis of FG-CNTMEE plates considering the effect of various circuit conditions associated with parameters such as CNT dis-tributions, volume fraction, skew angle, aspect ratio, length-to-thickness ratio and coupling fields has been carried out. The results of this article can serve as benchmark for future development and analysis of smart structures.     

基于改进核极限学习机和集成学习理论的目标机动轨迹预测

为了提高目标轨迹预测的精度以及预测模型的泛化能力,提出基于改进蝙蝠算法优化的核极限学习机(Kernel Extreme Learning Machine,KELM)和集成学习理论目标机动轨迹预测模型。构建KELM模型,并采用改进的蝙蝠算法对KELM的参数进行优化;以优化后的KELM神经网络为弱预测器,结合集成学习算法生成强预测器,通过训练不断优化强预测的结构和参数,得到一种基于集成学习理论的目标机动轨迹预测模型;基于不同规模的样本,将所得预测模型与逆传播神经网络、支持向量机和极限学习机等模型进行对比分析。仿真结果表明:所提目标机动轨迹预测模型具有较好的预测精度和泛化能力。