基于周期谱相关的扩频信号时差估计

周期谱相关时差估计法具有较强的抑制噪声和干扰的能力,适应于低信噪比的情况,但其抗噪声和干扰能力完全在于循环频率的选择.提出了运用周期谱特性选择扩频信号循环频率的方法,大大降低了谱相关时差估计算法的计算量.仿真结果表明,该方法是切实可行的.     

航天器环境干扰力矩的闭环辨识与补偿

研究引入干扰力矩辨识的方法提高三轴稳定航天器在稳态运行期间的高精度指向控制问题.基于航天器的稳态输出数据和系统传递函数,对环境干扰力矩的完整模型进行闭环辨识.其中采用改进的特征系统实现相关算法和多元线性回归法辨识环境干扰力矩的模型参数,最后使用前馈控制对所辨识的环境干扰力矩实现在线补偿.分析和仿真结果表明,该方案能有效地克服不确定的环境干扰力矩对姿态稳定度的影响,提高在轨运行航天器的姿态稳定精度.     

空间激光通信系统光斑小目标跟踪算法研究

在空间激光通信链路中,大气湍流、结构设计误差、平台扰动等因素为链路的精确对准和精密跟踪带来了困难。为提高激光链路的跟踪精度,首先对光斑跟踪过程中的影响因素进行研究分析,继而对跟踪算法做出改进。针对跟踪目标特性,设计了一种自适应模板目标相关跟踪算法,并搭建了基于粗精复合轴跟踪系统的实验平台,开展了实验验证。结果表明：与当前常用跟踪算法相比,该设计处理速度可达 1 kfps,目标跟踪位置与真实位置基本重合,目标跟踪准确率高于98%,在处理速度、跟踪准确率与算法鲁棒性上均有较大提升,具有一定的实用性。     

星载相控阵GNSS-R测高系统设计与实验

星载GNSS-R测高系统采用多波束相控阵天线和干涉互相关处理方式,接收北斗导航卫星和GPS导航卫星发射的L波段导航直射信号与海面反射信号,进行海面高度探测。详细介绍GNSS-R测高系统设计与实现、集成和外场试验等,外场试验不仅验证了相控阵GNSS-R测高系统功能,更重要的是验证了核心软件算法的正确性。GNSS-R测高系统适用于中尺度海洋现象的观测,可在时间和空间尺度上扩展雷达高度计的观测能力,并与之相互补充,有利于对复杂海面的中尺度结构进行较高时间分辨率的观测。GNSS-R测高仪为星载高精度海面高度探测提供了新型载荷。     

基于小波包变换及相关系数法的复合材料层合板冲击位置识别研究

复合材料较为广泛应用于航空、航天等工程领域,但对冲击载荷十分敏感。因此,对复合材料结构承受的冲击载荷进行在线监测以及冲击位置的实时识别具有重要意义。文章以复合材料层合板为研究对象,基于两个冲击位置的距离越靠近则接收到信号幅频特性相似度越高的特点,采用FBG光纤光栅传感器,通过小波包变换的方法来提取能量特征向量,同时结合相关系数法来实现复合材料层合板的冲击位置识别。在480 mm×480 mm的复合材料层合板上开展冲击实验,8次实验皆完成了冲击位置识别,其中7个点距离误差为0 mm,实现精准识别,另一个点误差在6%以内。     

基于实测数据的海杂波时空相关性分析

海杂波的相关特性分为时间相关性和空间相关性,二者均与雷达系统参数、环境条件等多种因素有关。文章利用X波段雷达实测海杂波数据,重点研究了不同极化条件下海杂波的时间相关性、高低典型海况和不同入射余角条件下的海杂波距离向和方位向空间相关性。经大量实测数据验证表明,极化方式对海杂波强时间相关性影响较小,海杂波距离向空间相关性受海况影响较大、方位向相邻分辨单元空间相关性较弱,这些结论对于海杂波中目标检测方法优化设计具有重要意义。     

Spatial approaches to moons from resonance relative to invariant manifolds

In this study, the final approach to a moon or other body from resonance is explored and compared to the invariant manifolds of unstable periodic orbits. It is shown that the stable manifolds of planar Lyapunov orbits can act as a guide for the periods or resonances that are required for the final approach in both the planar and spatial problems. Previously developed techniques for the planar problem are expanded for use with resonances and used for comparison with trajectories approaching a moon from these resonances. A similar technique is then used for exploring the relationship of invariant manifolds to approach trajectories in the spatial problem. It is shown that the invariant manifolds of unstable periodic orbits provide insight into the trajectory design, and they can be used as a guide to the more direct approach trajectories.     

Dynamics of a spacecraft and normalization around Lagrangian points in the Neptune–Triton system

The problem of a spacecraft orbiting the Neptune–Triton system is presented. The new ingredients in this restricted three body problem are the Neptune oblateness and the high inclined and retrograde motion of Triton. First we present some interesting simulations showing the role played by the oblateness on a Neptune’s satellite, disturbed by Triton. We also give an extensive numerical exploration in the case when the spacecraft orbits Triton, considering Sun, Neptune and its planetary oblateness as disturbers. In the plane a × I (a = semi-major axis, I = inclination), we give a plot of the stable regions where the massless body can survive for thousand of years. Retrograde and direct orbits were considered and as usual, the region of stability is much more significant for the case of direct orbit of the spacecraft (Triton’s orbit is retrograde). Next we explore the dynamics in a vicinity of the Lagrangian points. The Birkhoff normalization is constructed around L₂, followed by its reduction to the center manifold. In this reduced dynamics, a convenient Poincaré section shows the interplay of the Lyapunov and halo periodic orbits, Lissajous and quasi-halo tori as well as the stable and unstable manifolds of the planar Lyapunov orbit. To show the effect of the oblateness, the planar Lyapunov family emanating from the Lagrangian points and three-dimensional halo orbits are obtained by the numerical continuation method.     

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%。     

Investigations into sensing characteristics of circular thin-plate electrostatic sensors for gas path monitoring

Circular thin-plate electrostatic sensors are promising in gas path monitoring due to their advantages of non-intrusiveness and easy installation. The spatial sensitivity and filtering effect are two important performance parameters. In this paper, an analytically mathematical model of induced charge on a circular thin-plate electrode is first derived. Then the spatial sensitivity and filtering effect of the circular electrostatic sensor are investigated by numerical calculations. Finally,experimental studies are performed to testify the theoretical results. Both theoretical and experimental results demonstrate that circular thin-plate electrostatic sensors act as a low-pass filter in the spatial frequency domain, and both the spatial filtering effect and the temporal frequency response characteristics depend strongly on the spatial position and velocity of the charged particle. These conclusions can provide guidelines for the optimal design of circular thin-plate electrostatic sensors.