对单脉冲雷达角度跟踪系统的干扰仿真研究

在介绍单脉冲雷达角度跟踪系统原理的基础上,从雷达回波信号的角度建立了两相干干扰源对单脉冲雷达角度跟踪系统进行干扰的具体模型,给出了仿真流程,详细分析了目标雷达波束指向角、两干扰源与目标雷达夹角、两干扰源相位差以及干扰功率比等因素对干扰效果的影响。     

Crew autonomy for deep space exploration: Lessons from the Antarctic Search for Meteorites

Future piloted missions to explore asteroids, Mars, and other targets beyond the Moon will experience strict limitations on communication between vehicles in space and control centers on Earth. These limitations will require crews to operate with greater autonomy than any past space mission has demonstrated. The Antarctic Search for Meteorites (ANSMET) project, which regularly sends small teams of researchers to remote parts of the southern continent, resembles a space mission in many ways but does not rely upon a control center. It provides a useful crew autonomy model for planners of future deep space exploration missions. In contrast to current space missions, ANSMET gives the crew the authority to adjust competing work priorities, task assignments, and daily schedules; allows the crew to be the primary monitor of mission progress; demands greater crew accountability for operational errors; requires the crew to make the most of limited communication bandwidth; adopts systems designed for simple operation and failure recovery; and grants the crew a leading role in the selection and stowage of their equipment.     

自然通风和蒸发冷却在新疆建筑应用的数值分析

为了使新疆和田地区夏季在经济性合理前提下使室内通风环境满足人体热舒适要求。利用蒸发冷却技术将室外空气处理后送入室内，通过数值模拟分析得出在夏季室内负荷最大时刻采用机械辅助式自然通风，使得室温能够雏持在26-8℃，风速为0．6m/s，基本满足人体热舒适要求。在室外温度适宜的过渡季节采用Trombe墙和太阳能烟囱复合的自然通风系统，建筑太阳能烟囱的经济高度为10m。     

圆球绕流场的尾涡分析和升阻力研究

采用自编的分块算法程序模拟了雷诺数在20～1000之间的圆球绕流场.在Re=25时捕捉到流动分离,与试验结果一致.对Re=250时稳态非对称尾流区内流体的输运情况进行研究,证实圆球绕流场中纵向对称面的存在.对于非稳态流动,主要研究了阻力和横向力的时间变化规律,并分析了频谱特征.计算发现横向合力主要在横截面上两个很窄的区间内偏移.经过几种不同涡定义方法的优劣测试,本文采用Hunt等人介绍的Q-定义方法对非稳态绕流场的尾涡结构进行了直观描述.     

Space science education in Egypt and the 2006 solar eclipse

The space science research has been started in Egypt since 1910 by measuring the solar constant as indication of solar radiation at Helwan Observatory. The solar sunspot studies and its influence on the Nile flooding was erected and operated at Helwan as a first solar station in Egypt during 1957. Zeiss-Coude’ refractor was installed in 1964. Astronomy and space science educations started in Egypt at the university level since 1936 at Department of Astronomy and Meteorology of Cairo University. Undergraduate and graduate education in Egypt will be discussed in this work.     

关于Access2000的安全机制解析

Microsoft Access 2000数据库作为常用的桌面数据库软件，在中小型的数据库系统中得到了广泛的应用。针对Access2000安全性能不高，探讨如何设置及提高Access2000的安全性方法。     

针栓式喷注单元膜束撞击雾化混合过程数值模拟

为了全面认识针栓式喷注器喷雾场结构，基于自适应网格加密技术和分三相计算的PLIC VOF（Piecewise Linear Interface Calculation Volume of Fluid）方法对针栓式喷注单元膜束撞击雾化混合过程进行了仿真分析，通过对两路推进剂分别进行界面追踪，获得了膜束撞击雾化混合过程的详细结构特征，与高速摄影试验结果定性定量对比均吻合较好，验证了数值方法的准确性。以此为基础对膜束撞击的喷雾场结构、撞击变形过程、流场涡结构、雾化破碎典型特征及破碎后的雾化混合分布特征进行了识别分析，结果表明：膜束撞击形成了液束未穿透液膜和液束穿透液膜2种不同的喷雾扇结构。膜束撞击形成的喷雾扇呈"Ω"形，膜束同时发生弯曲变形和横截面变形。另外，膜束撞击同时受到正压和剪切应力作用，导致了一系列复杂涡流现象，使得相互作用增强，雾化混合均增强，这也是膜束撞击喷注构型优于膜膜撞击的本质原因。最后，还发现膜束撞击喷雾场液滴分布呈现分区结构特征，分别是液束控制主导的上雾化区、液膜控制主导的下雾化区及夹在中间的混合区，实际中应兼顾雾化特性和混合特性，选取中等动量比膜束撞击，这可为针栓式喷注器的理论研究和工程设计提供重要参考。     

基于自适应深度网络的无人机目标跟踪算法

针对无人机（UAV）视频中目标易受到遮挡、形变、复杂背景干扰等问题，提出一种基于自适应深度网络的无人机目标跟踪算法。首先，基于主成分分析（PCA）和卷积神经网络（CNN）算法，设计3阶的自适应深度网络进行目标特征提取，该网络对图像的H、S、I通道分别进行主成分分析学习，将得到的特征向量输入网络进行分层卷积，优化了网络结构，提高了网络的收敛速度和精度。其次，将目标深度特征输入核相关滤波算法进行目标跟踪，通过分析相邻2帧图像的变化率，采用分段自适应调整学习率的算法进行目标模板更新，有效地改善目标遮挡问题。仿真实验结果表明，该算法有效地避免了复杂因素干扰导致的跟踪精度下降，具有较好的鲁棒性，相较于全卷积跟踪（FCNT）算法平均跟踪精度提高了9.62%，平均跟踪成功率提高了11.9%。     

由感知到动作决策一体化的类脑导航技术研究现状与未来发展

随着脑与神经科学以及人工智能技术的持续发展，昆虫和哺乳动物大脑导航机理启发下的感知/认知/路径规划/动作决策一体化类脑导航技术得到了较大发展，可以实现由原始感知信息输入到导航动作决策的直接输出，呈现出接近动物端到端面向目标导航的智能行为，具有提高密集型无人机集群导航鲁棒性、准确性、实时响应动作、自主智能性以及计算效率的潜力。阐述了昆虫和哺乳动物大脑导航机理及其互补对称关系，以及昆虫和哺乳动物大脑导航机理启发的端到端类脑导航技术内涵；论述了类脑导航技术研究进展，包括类脑环境感知、类脑空间认知、面向目标类脑导航；分析了类脑导航向智能化、神经形态系统以及群体导航发展的新趋势；最后讨论了类脑导航技术应用于无人机密集集群系统时存在的挑战。     

Route generation and description using the notions of object's influence area and spatial conceptual map

In the GRAAD Project we aimed at creatinga system which could generate route directions thatare comparable to route directions created by humanparticipants. With this goal in mind, we started froma linguistic and cognitive study of route directionsproduced by people and the study of cognitive modelsof mental maps. We proposed a new qualitative spatialmodel that can support the spatial properties of humanroute directions. This model is based on the notion ofobject's influence area which is used to modelneighborhood, orientation and distance. The proposedapproach relies on the manipulation of spatialentities in a spatial conceptual map (SCM) which isthe computarized analog of a mental map used bypeople. We developped the GRAAD System, software thatgenerates routes in a SCM and describes them innatural language. Finally, we conducted an experimentin order to compare GRAAD's route directions androutes described by human participants in similarexperimental conditions. GRAAD's output was notdistinguishable from route directions created by humanparticipants. In this paper we present the mainresults obtained during all phases of the GRAADProject.