鲁棒EKF在脉冲星导航系统中的应用

针对脉冲星导航系统的滤波问题,传统的扩展卡尔曼滤波(EKF)算法存在不能克服系统模型存在不确定性参数以及乘性噪声等缺陷,提出一种鲁棒EKF算法。首先,分析了状态预测误差方程和估计误差方程,利用统计学原理,得到了状态预测方差矩阵和状态估计方差矩阵计算等式。由于系统模型存在不确定性参数,状态预测协方差矩阵和状态估计协方差矩阵无法计算;因此,利用4个重要矩阵不等式,分析并找到预测方差矩阵和状态估计方差矩阵的上界。最后,利用状态估计误差协方差矩阵上界设计状态增益矩阵,使得状态估计协方差矩阵的迹最小。将该算法对脉冲星导航系统进行仿真,仿真结果验证了所提算法的有效性。     

状态噪声与测量噪声方差的自适应序贯估计

本文将Jazwinski的状态噪声方差的估计推广到测量系统是多维的、状态噪声协方差矩阵不再是纯量,而是对角矩阵情况下的状态噪声方差的估计,同时可得到测量噪声协方差阵和状态向量估计,从而形成自适应滤波。     

基于GPR模型的自适应平方根容积卡尔曼滤波算法

与传统算法一样,动态系统的参数化模型(含噪声统计特性)未知或不够准确易导致容积卡尔曼滤波(CKF)效果严重下降,甚至滤波结果发散.为此,利用高斯过程回归(GPR)方法对训练数据进行学习,得到动态系统的状态转移GPR模型和量测GPR模型以及噪声统计特性,用以替代或增强原有动态系统模型,并将其融入到平方根容积卡尔曼滤波(SRCKF)中,分别提出了无模型高斯过程SRCKF (MFGP-SRCKF)和模型增强高斯过程SRCKF (MEGP-SRCKF)两种算法.仿真结果表明:这两种新的自适应滤波算法提高了动态系统模型精度,且实时自适应调整噪声的协方差,克服了传统算法滤波性能易受系统模型限制的问题;与MFGP-SRCKF相比,在给定一个不够准确的参数化模型,且有限的训练数据未能遍布估计状态空间的情况下,MEGP-SRCKF具备更高的滤波精度.     

基于自适应容积卡尔曼滤波方法的涡扇发动机气路部件故障诊断

针对涡扇发动机气路部件故障诊断中参数存在不同的噪声统计特性,提出了一种自适应平方根容积卡尔曼滤波(ASRCKF)器的自适应滤波方法.该方法直接利用基于3阶容积积分方法近似发动机的非线性统计特性,用于替代非线性无迹卡尔曼滤波方法的系统模型,避免了滤波过程参数选取的问题;采用移动窗口法对噪声协方差矩阵进行自适应估计,提高了算法对不同统计特性噪声的自适应能力和滤波精度.通过对发动机气路部件健康参数蜕化过程仿真结果表明:ASRCKF方法相比平方根容积卡尔曼滤波(SRCKF)方法,精度提高40%~50%,对不同噪声信号具有更好的适应能力.      

基于PnP和自适应线性卡尔曼滤波的实时位姿估计

针对实时位姿估计中扩展卡尔曼滤波(EKF)线性化引入非线性误差和依赖已知噪声分布的缺点,提出一种基于PnP的自适应线性卡尔曼滤波位姿估计求解方法。将PnP位姿估计求解策略引入卡尔曼滤波观测方程,通过对动态方程误差统计参数实时估计,自适应调节卡尔曼滤波递推参数。所提算法求解精度高,固定了观测方程的观测向量维度,提高了算法实用性。通过仿真试验,比较了该算法与EKF的位姿估计精度,通过量化误差分析,证明了该方法可以提高三维运动位姿估计精度,也验证了该方法的有效性。     

基于改进平方根无迹卡尔曼滤波方法的涡扇发动机气路状态监控

针对涡扇发动机气路状态监控存在模型未知或不准确导致滤波效果下降甚至发散的问题,研究了一种融入高斯过程回归(GPR)的改进平方根无迹卡尔曼滤波(UKF)方法.该方法利用GPR对训练数据进行学习,建立发动机气路部件状态监控的GPR模型,替代UKF方法中的非线性系统模型;采用超球体单形采样和平方根滤波方法来提高滤波的计算效率和数值稳定性.仿真结果表明:训练的GPR模型解决了UKF方法对发动机原系统模型和噪声协方差矩阵依赖性的问题;与扩展卡尔曼滤波(EKF)和平方根UKF方法相比较,改进平方根UKF方法精度更高,对健康参数的估计精度达到99.9%,实现了对涡扇发动机单个和多个气路部件健系参数的有效跟踪.     

基于先验知识及其定量评估的自适应杂波抑制研究

 为了改善训练样本数受限的非均匀杂波环境中的系统检测性能,研究了基于先验知识及其定量评估的自适应杂波抑制算法。提出了使用经真实杂波信息白化后的先验杂波协方差矩阵与单位矩阵之差的谱范数,来定量评估杂波先验知识的准确程度,并给出了真实杂波协方差矩阵未知时的矩阵谱范数估计方法。结合先验知识定量评估结果,获得了具有先验知识约束时的杂波协方差矩阵最大似然估计方法。分别基于多脉冲相参雷达以及空时自适应雷达进行了杂波建模,在此基础之上分析了算法性能。仿真结果证实了该算法优于使用样本协方差矩阵及先验杂波信息形成杂波抑制权值的性能。     

基于鲁棒自适应状态观测器的异步电机磁链估计抗扰性研究

针对传统全阶磁链观测器随内、外扰动导致磁链估测精度不高甚至系统不稳定的问题,对观测器极点分布情况和内、外扰动对系统的影响进行研究分析,从而对反馈矩阵增益系数的取值进行设计,提出了一种鲁棒自适应状态观测器对磁链进行估计。当遇到内、外扰动时,在保持系统稳定的情况下自动减小反馈增益系数以减小扰动对系统的影响,从而保持系统良好的动静态性能。在内、外扰动下,将鲁棒自适应状态观测器与传统的全阶磁链观测器的磁链估计精度、转速波动以及转矩波动进行比较分析。仿真和试验结果表明:鲁棒自适应状态观测器具有比传统全阶磁链观测器更好的抗干扰性能。     

自适应渐消Kalman滤波算法在RTK中的应用研究

在动态载波相位差分定位(RTK)中,由于观测环境复杂,会经常发生周跳、卫星信号失锁等情况,严重影响基线解算的连续性和可靠性。针对动态应用环境,提出了一种Kalman滤波算法在RTK技术中的应用方法。该方法可以实时估计模糊度浮点解及其协方差矩阵,在需要重新固定模糊度时可直接用于搜索,起到了周跳修复的作用。此外,采用了自适应渐消Kalman滤波算法提高算法的动态适应性,并引入独立的滑动窗进行新息的收集和处理,解决了由于参考星变化或卫星信号失锁造成观测量中断而无法准确计算新息协方差的难题。仿真结果表明,该算法能够在模糊度发生变化时快速收敛,并且相对于一般Kalman滤波算法在高动态下提高了模糊度浮点解的精度,提高了后续模糊度搜索的效率和固定成功率。     

非高斯杂波下自适应雷达目标检测新方法

在球不变随机向量(SIRV)非高斯杂波背景下,研究了多脉冲相参雷达目标的自适应检测问题。假设杂波具有相同的协方差矩阵结构和可能相关的纹理分量,提出了新的协方差矩阵估计器,并获得了相应的自适应归一化匹配滤波器(ANMF)。理论分析表明,在估计杂波分组大小与实际情况匹配时,所获得的ANMF对杂波功率水平和协方差矩阵结构均具有恒虚警率(CFAR)特性。仿真结果表明:当估计的杂波分组大小失配时,所获得的ANMF具有近似CFAR特性,并进一步分析了不同参数变化对所提检测器性能的影响。与已有的ANMF相比,所获得的ANMF具有更好的检测性能,且迭代次数更小,其相对于已知杂波协方差矩阵的最优归一化匹配滤波器(NMF)的检测损失也更小,具有很好的实际应用前景。     

EMD-EKF方法研究

应用扩展卡尔曼滤波(EKF)时需要估计量测噪声的统计特性。文中针对观测噪声统计特性描述不准确导致的EKF性能下降的问题,利用经验模态分解方法(Empirical Mode Decomposition,EMD)可以分离信号和噪声的特性,提出了一种在未知量测噪声条件下的EKF方法。该方法可以跟踪观测噪声的变化,即实现了对量测噪声的估计,从而解决了在未知量测噪声的情况下的EKF问题。仿真结果表明可运用于无源定位中。     

跟飞编队卫星相对导航自适应EKF算法研究

针对跟踪星对目标星跟飞编队的相对导航问题,提出了基于自适应扩展卡尔曼滤波（EKF）的相对导航算法。以线性离散化的椭圆参考轨道相对动力学模型为导航状态方程．设计了虚拟测量量及相应的测量矩阵,避免了求解雅可比矩阵的复杂计算。为了适应构型尺寸变化引起的模型误差变化,提出了模型误差在线估计算法。仿真显示,算法具有较快的收敛速度和较高的估计精度。该研究成果可作为卫星编队应用的有益参考。     

包含乘性噪声自适应修正的非合作目标相对导航算法

无人机（UAV）态势感知的任务是利用机载传感器对未知环境进行目标识别和引导，针对无人机与非合作目标间中远距离的相对导航问题，提出了一种基于角度和距离量测的相对状态估计算法。在现有滤波算法的基础上，为了提高精度和稳定性，本文利用了列文伯格-马夸尔特（LM）优化的思想对迭代卡尔曼滤波（IEKF）算法进行改进，提出了一种LM-IEKF算法，并推导该算法在迭代过程中的状态更新方程及协方差阵的递推公式。在此基础上，考虑到距离传感器由于信号相关特性而引入的乘性噪声，现有的加性噪声模型难以适应，因此，进一步提出了基于量测噪声自适应修正的Modified LM-IEKF方法，通过在线实时更新噪声阵提高滤波的精度，并设置渐消记忆指数平滑估计结果。算法验证结果表明，与现有的EKF、IEKF算法相比，在仅含加性噪声的情况下，LM-IEKF算法具有更好的性能；在包含乘性噪声的情况下，Modified LM-IEKF可以有效地估计量测噪声，与目前广泛使用的EKF算法相比，在综合相对位置和相对速度精度上分别提高了10%和23%。     

Rapid Convergence Rate in Adaptive Arrays

In many applications, the practical usefulness of adaptive arrays is limited by their convergence rate. The adaptively controlled weights in these systems must change at a rate equal to or greater than the rate of change of the external noise field (e.g., due to scanning in a radar if step scan is not used). This convergence rate problem is most severe in adaptive systems with a large number of degrees of adaptivity and in situations where the eigenvalues of the noise covariance matrix are widely different. A direct method of adaptive weight computation, based on a sample covariance matrix of the noise field, has been found to provide very rapid convergence in all cases, i.e., independent of the eigenvalue distribution. A theory has been developed, based on earlier work by Goodman, which predicts the achievable convergence rate with this technique, and has been verified by simulation.     

Robust adaptive filtering method for SINS/SAR integrated navigation system

This paper presents a new robust adaptive filtering method for SINS/SAR (Strap-down Inertial Navigation System/Synthetic Aperture Radar) integrated navigation system. This method adopts the principle of robust estimation to adaptive filtering of observational data. A robust adaptive filter is developed to adaptively determine the covariance matrix of observation noise, and adaptively adjust the covariance matrix of system state noise according to the adaptive factor constructed based on predicted residuals. Experimental results and comparison analysis demonstrate that the proposed method cannot only effectively resist disturbances due to system state noise and observation noise, but it can also achieve higher accuracy than the adaptive Kalman filtering method.     

A new SINS/GPS sensor fusion scheme for UAV localization problem using nonlinear SVSF with covariance derivation and an adaptive boundary layer

The state estimation strategy using the smooth variable structure filter(SVSF) is based on the variable structure and sliding mode concepts. As presented in its standard form with a fixed boundary layer limit, the value of the boundary layer width is not precisely known at each step and may be selected based on a priori knowledge. The boundary layer width reflects the level of uncertainty in the model parameters and disturbance characteristics, where large values of the boundary layer width lead to robustness without optimality and small values of the boundary layer width provide optimality with poor robustness. As a solution and to overcome these limitations, an adaptive smoothing boundary layer is required to achieve greater robustness and suitable accuracy.This adapted value of the boundary layer width is obtained by minimizing the trace of the a posteriori covariance matrix. In this paper, the proposed new approach will be considered as another alternative to the extended Kalman filters(EKF), nonlinear H1 and standard SVSF-based data fusion techniques for the autonomous airborne navigation and self-localization problem. This alternative is based on strapdown inertial navigation system(SINS) and GPS data using the nonlinear SVSF with a covariance derivation and adaptive boundary layer width.Furthermore, the full mathematical model of the SINS/GPS navigation system considering the unmanned aerial vehicle(UAV) position, velocity and Euler angle as well as gyro and accelerometer biases will be used in this paper to estimate the airborne position and velocity with better accuracy.     

Artificial Noise in Adaptive Arrays

"Artificial noise," or the connection of feedback paths around the the integrators, is shown to be an effective method of dealing with the problem of multiplier offsets in adaptive antennas. This probl which was analyzed by Compton [1] is particularly troubles when the covariance matrix is singular or nearly so. Like added real noise, the artificial noise improves the condition number of the underlying matrix. The artificial noise, however, avoids the obvious disadvantage of adding to the real noise level. As a result the output-signal-to-interference ratio is much less degraded.     

Reduced-rank STAP performance analysis

The space-time radar problem is well suited to the application of techniques that take advantage of the low-rank property of the space-time covariance matrix. It is shown that reduced-rank (RR) methods outperform full-rank space-time adaptive processing (STAP) when the space-time covariance matrix is estimated from a data set with limited support. The utility of RR methods is demonstrated by theoretical analysis, simulations and analysis of real data. It is shown that RR processing has two opposite effects on the performance: increased statistical stability which tends to improve performance, and introduction of a bias which lowers the signal-to-noise ratio (SNR). A method for evaluating the theoretical conditioned SNR for fixed RR transforms is also presented. It Is shown that while best performance is obtained using data-dependent transforms, the loss incurred by the application of fixed transforms (such as the discrete cosine transform) may be relatively small. The main advantage of fixed transforms is the availability of efficient computational procedures for their implementation. These findings suggest that RR methods could facilitate the development of practical, real-time STAP technology     

Hierarchical adaptive Kalman filtering for interplanetary orbitdetermination

A modular and flexible approach to adaptive Kalman filtering has recently been introduced using the framework of a mixture-of-experts regulated by a gating network. Each expert is a Kalman filter modeled with a different realization of the unknown system parameters. The unknown or uncertain parameters can include elements of the state transition matrix, observation mapping matrix, process noise covariance matrix, and measurement noise covariance matrix. The gating network performs on-line adaptation of the weights given to individual filters based on performance. The mixture-of-experts approach is extended here to a hierarchical architecture which involves multiple levels of gating. The proposed architecture provides a multilevel hypothesis testing capability. The utility of the hierarchical architecture is illustrated via the problem of interplanetary navigation (Mars Pathfinder) using simulated radiometric data. It serves as a useful tool for assisting navigation teams in the process of selecting the parameters of the navigational filter over various operating regimes. It is shown that the scheme has the capability of detecting changes in the system parameters and switching filters appropriately for optimal performance. Furthermore, the expectation-maximization (EM) algorithm is shown to be applicable in the proposed framework