新生目标强度未知的双门限粒子PHD滤波器

传统粒子概率假设密度(PHD)滤波器假定新生目标强度已知,当新生目标在整个观测区域随机出现时不再适用。为解决新生目标强度未知时的多目标跟踪问题,提出了一种基于量测信息的双门限粒子PHD(PHD-DT)滤波器。首先基于似然函数设定门限对存活目标量测进行粗提取,利用上一时刻的目标估计值构建圆形波门进行精细提取,并对门限设定方法进行分析,然后根据提取结果对目标PHD进行分解,得到存活目标和新生目标的PHD预测及更新表达式,最后给出了滤波器的实现方法并同基于量测驱动的PHD(PHD-M)滤波器和Logic+联合概率数据互联(JPDA)方法进行了仿真对比。仿真结果表明,在新生目标强度未知时,PHD-DT可有效避免Logic+JPDA在杂波背景下因航迹起始错误带来的估计误差,并较好地解决了PHD-M的目标数目过估问题,多目标估计性能更优,且杂波越强性能优势越明显。     

失效卫星远距离相对位姿估计与优化方法

针对远距离非合作慢旋目标位姿估计精度问题,提出一种融合图像超分辨与视觉SLAM的相对位姿估计方法。算法主要包含3个步骤：通过梯度引导生成式对抗超分辨技术,提升目标图像的质量以获取更多更高质量的特征点;构建特征数据库实现当前帧与特征数据库的匹配,提升旋转目标的特征跟踪稳定性;利用图优化对多帧图像进行联合位姿优化,消除累计误差,得到更为精确的估计结果。为稳定网络的训练,将自然进化算法引入到对抗训练中。为增强模型的泛化性和鲁棒性,实验中的数据集采用半物理仿真获得。实验结果表明,当等效距离为25 m且失效卫星以25（°）/s的速度旋转时,目标图像经超分辨网络增强后,能够实现连续稳定的长时间测量。     

火星进入段探测器自校准状态估计

由于大气密度、气动参数、突风和沙尘暴等因素的影响,火星探测器在进入段高速飞行的动力学模型中往往带来未知输入,这些未知输入使传统的滤波方法出现较大的偏差。研究采用一种新的自校准扩展Kalman滤波方法,对火星进入段的探测器进行状态估计,可以成功地消除这些未知输入带来的影响。数值仿真结果表明,该方法能有效提高导航精度。     

空间绳系机器人抓捕非合作目标的质量特性参数辨识

针对空间绳系机器人抓捕非合作目标/空间垃圾后需要对其进行回收/拖曳的精确控制问题,提出了一种利用抓捕后保持阶段的振动特性辨识目标参数的方法。首先,根据质量特性参数辨识的需要,推导了系统的动力学模型。不同于以往将本体卫星和被抓捕目标简化为质点的动力学模型,本文针对任意的目标抓捕位置,在考虑重力梯度影响的基础上,利用拉格朗日法获得系统各广义坐标的动力学公式。然后,分析非合作目标和系绳在后抓捕保持阶段的姿态运动。最后,在非合作目标与本体卫星没有任何信息交互的情况下,利用后抓捕阶段目标卫星和系绳特有的振动,并使用具有鲁棒性可遗忘因子的递推最小二乘法,提出了包括转动惯量和质心到任意抓捕点距离在内的质量特性参数辨识算法。     

对空间碎片的相对位姿估计

针对几何、质量特性参数均未知的空间碎片,提出一种基于扩展卡尔曼滤波（EKF）的相对位姿估计算法。该算法以单目相机获得的目标图像特征视线信息作为测量输入,以目标姿态、姿态角速度、惯量比、相对位置、相对速度和特征位置作为滤波状态。对算法进行可观性分析,给出两个可以保证算法可观性的条件。分别针对不同条件进行数值仿真,仿真结果表明新算法能够对几何和质量参数未知的空间碎片进行有效的相对位姿 估计。     

Electrochemical trepanning with an auxiliary electrode

Electrochemical trepanning (ET) is one of the main methods for processing the blisks in aero-engines. However, stray corrosion in ET has a very negative impact on the machining quality of the blades. This paper proposes an innovative ET method with an auxiliary electrode surrounding the inner wall of the cavity above the cathode. Two-dimensional electric-field models are established and simulations are carried out with the auxiliary electrode at different positions and different electric potentials. The electric-field simulations show that adding the auxiliary electrode reverses the directions of the current lines in the processed area, thereby protecting it. The results show that when the distance from the bottom of the auxiliary electrode to the top of the cathode tool is 3 mm and the potential difference between the auxiliary electrode and the anode workpiece is 5 V, the area corroded by stray current is decreased. Experiments are performed under the parameter values determined by the simulations, and they confirm that this method is effective at reducing stray corrosion. Specifically, at a feed rate of 2.5 mm/min, the blade taper angle is decreased from 1.37° to −0.09°, the thickness of the leading and trailing edges is increased from 0.307 mm to 0.704 mm, and the average surface roughness is decreased from S_a =8.239 μm to S_a=7.028 μm.     

Broglio Drag Balance for neutral thermosphere density measurement on UNICubeSAT

The nanosatellite UNICubeSAT is described, carrying a Broglio Drag Balance Instrument for neutral thermosphere density in situ measurements. The aim of the mission is to contribute to the development of accurate thermosphere models, achieving in situ, real time measurements of atmosphere density, that could be exploited for global atmosphere model validation and accurate short term (1–3 days) real time space weather forecasts. The satellite is inexpensive and swarms could be easily launched operating as a distributed sensor network to get simultaneous in situ local (not orbit averaged) measurements in multiple positions and orbit heights. The nanosatellite is based on the Cubesat standard architecture, weighing about 1 kg for 1-L volume. Atmospheric drag force is measured by the displacement of light plates exposed to the incoming particle flux seen by the spacecraft, applying the original three dimensional Broglio Drag Balance concept to a single nanosatellite axis. The instrument concept and its relation to the satellite bus is depicted, showing that many long term potential measurement error sources and biases can be removed in data processing if the spacecraft is spin stabilized. The expected accuracy in density measurements is 20%. The instrument cost is a fraction of that of accurate accelerometers. The onboard systems are based on commercial off the shelf components, in accordance with the short lifetime typical of aeronomy satellites.     

基于STLS的卫星惯量矩阵在轨估计

提出了一种采用结构总体最小二乘（Structured total least squares, STLS）进行卫星惯量矩阵在轨估计的方法,与当前估计方法相比,该方法在考虑敏感器测量噪声时能获得一致估计。首先由动量守恒定律得到估计方程,针对该方程的特点定义了惯量矩阵的STLS估计,并使用结构总体最小范数（Structured total least norm, STLN）算法进行求解。证明了当噪声为高斯分布时该STLS估计为极大似然估计,给出了该STLS估计具有一致性的充分条件,仿真结果验证了文章所提估计方法的有效性。     

非质心特征点相对位置和姿态估计方法研究

卫星编队飞行中,从星星体上偏离质心的任意特征点,其相对位置和相对姿态运动耦合。本文针对任意点的相对状态估计问题,推导了耦合相对轨道动力学方程和无陀螺相对姿态动力学方程,并且基于PSD敏感器对主星上点光源的视线测量值,提出了基于特征点耦合动力学和质心传统动力学的两种相对状态估计策略。针对每种策略分别设计了EKF和UKF两种滤波算法对相对状态进行估计。最后通过数学仿真对算法的有效性进行了验证,并对其计算量进行了分析。结果表明,四种算法均能对特征点的相对状态进行较高精度的估计。     

图解建模方法在提取沟壑密度中的分析研究

沟壑密度是反映地表破碎程度的重要因子,在土壤侵蚀、地貌演变等研究中有重要作用。对于沟壑密度的提取方法目前较为复杂,这主要由于在提取过程中所涉及到的因子数目众多,为进一步的空间分析研究设置了障碍。利用图解建模分析方法建立一套沟壑密度提取分析空间数据模型,对沟壑密度的快速提取提供了一套全新的解决方案。最后,以四川省数据为例,采用该分析方法提取出了四川省沟壑密度数据,为进一步研究四川省地形地貌特征奠定了基础。