不确定重尾量测噪声干扰下的鲁棒目标跟踪算法

在大多数目标跟踪方法中,通常假设量测噪声服从参数已知的高斯分布或对称重尾分布,但其非常受限并且在实际过程中常常无法得到满足。因此,针对存在不确定重尾量测噪声干扰下的目标跟踪问题,提出基于变分推理的鲁棒容积卡尔曼滤波算法。该算法利用Skew-T分布对不确定重尾量测噪声进行建模,在基于容积规则的数值积分过程中,结合变分推理将Skew-T分布量测噪声参数与系统状态变量进行联合递归计算,通过对近似后验概率密度函数进行变分迭代,获得系统模型和不对称重尾量测噪声参数。仿真结果表明,该算法相比变分贝叶斯扩展卡尔曼滤波算法具有较高的滤波精度。     

Maneuvering target tracking algorithm based on CDKF in observation bootstrapping strategy

The selection and optimization of model filters affect the precision of motion pattern identifica-tion and state estimation in maneuvering target tracking directly.Aiming at improving performance of model filters, a novel maneuvering target tracking algorithm based on central difference Kalman filter in observation bootstrapping strategy is proposed.The framework of interactive multiple model ( IMM) is used to realize identification of motion pattern, and a central difference Kalman filter ( CDKF) is selected as the model filter of IMM.Considering the advantage of multi-sensor fusion method in improving the stability and reliability of observation information, the hardware cost of the observation system for multiple sensors is adopted, meanwhile, according to the data assimilation technique in Ensemble Kalman filter( EnKF) , a bootstrapping observation set is constructed by in-tegrating the latest observation and the prior information of observation noise.On that basis, these bootstrapping observations are reasonably used to optimize the filtering performance of CDKF by means of weight fusion way.The object of new algorithm is to improve the tracking precision of ob-served target by the multi-sensor fusion method without increasing the number of physical sensors. The theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.     

Maneuvering target tracking algorithm based on cubature Kalman filter with observation iterated update

Reasonable selection and optimization of a filter used in model estimation for a multiple model structure is the key to improve tracking accuracy of maneuvering target.Combining with the cubature Kalman filter with iterated observation update and the interacting multiple model method,a novel interacting multiple model algorithm based on the cubature Kalman filter with observation iterated update is proposed.Firstly,aiming to the structural features of cubature Kalman filter,the cubature Kalman filter with observation iterated update is constructed by the mechanism of iterated observation update.Secondly,the improved cubature Kalman filter is used as the model filter of interacting multiple model,and the stability and reliability of model identification and state estimation are effectively promoted by the optimization of model filtering step.In the simulations,compared with classic improved interacting multiple model algorithms,the theoretical analysis and experimental results show the feasibility and validity of the proposed algorithm.     

改进的基于奇异值分解的抗差容积卡尔曼滤波算法在全球定位导航中的应用

非线性动态系统存在非线性和噪声不确定的问题,容积卡尔曼滤波对解算该类系统有较好的精度,为了提升导航系统对异常观测值的稳定性,对采样数据进行均值滤波处理,降低干扰较大的采样数据对于滤波结果的影响.用奇异值分解代替Cholesky分解,改善滤波稳定性,避免先验协方差非正定而降低滤波性能.最后通过引入抗差因子调节观测协方差矩阵,再次减少观测异常值对于滤波结果的影响.采用仿真实验进行分析,改进的抗差容积卡尔曼滤波算法对于减弱异常观测值影响有良好的效果.     

Marginalized cubature Kalman filtering algorithm based on linear/nonlinear mixed-Gaussian model

Aiming at improving the estimation accuracy and real-time of nonlinear system with linear Gaussian sub-structure,a novel marginalized cubature Kalman filter is proposed in Bayesian estimation framework. Firstly,the marginalized technique is adopted to model the target system dynamics with nonlinear state and linear state separately,and the two parts are estimated by cubature Kalman filter and standard Kalman filter respectively. Therefore,the linear part avoids the generation and propagation process of cubature points. Accordingly,the computational complexity is reduced.Meanwhile,the accuracy of state estimation is improved by taking the difference of nonlinear state estimation as the measurement of linear state. Furthermore,the computational complexity of marginalized cubature Kalman filter is discussed by calculating the number of floating-point operation. Finally,simulation experiments and analysis show that the proposed algorithm can improve the performance of filtering precision and real-time effectively in target tracking system.     

自适应强跟踪容积卡尔曼滤波算法

提出了一种自适应强跟踪容积卡尔曼滤波算法(ASTSCKF),该算法在平方根容积卡尔曼滤波算法(SCKF)步骤中引入强跟踪滤波器(STF),通过渐消因子在线修正一步预测误差协方差矩阵,强迫输出残差序列正交,使得算法具有应对系统状态突变等不确定因素的能力,增强了算法的鲁棒性;结合改进渐消记忆时变噪声统计估计器,对噪声方差阵进行实时在线估计,有效解决了SCKF算法由于噪声统计不准确、未知或时变性带来的滤波发散问题,使其具有应对噪声变化的自适应能力。仿真实验结果表明：ASTSCKF算法在系统状态发生突变并且噪声变化的情况下,能够表现出良好的滤波性能,较SCKF算法有更强的鲁棒性以及噪声变化的自适应性。     

Cubature Kalman probability hypothesis density filter based on multi-sensor consistency fusion①

The GM-PHD framework as recursion realization of PHD filter is extensively applied to multi-target tracking system .A new idea of improving the estimation precision of time-varying multi-target in non-linear system is proposed due to the advantage of computation efficiency in this paper .First, a novel cubature Kalman probability hypothesis density filter is designed for single sensor measure -ment system under the Gaussian mixture framework .Second , the consistency fusion strategy for multi-sensor measurement is proposed through constructing consistency matrix .Furthermore, to take the advantage of consistency fusion strategy , fused measurement is introduced in the update step of cubature Kalman probability hypothesis density filter to replace the single-sensor measurement .Then a cubature Kalman probability hypothesis density filter based on multi-sensor consistency fusion is proposed .Capabilily of the proposed algorithm is illustrated through simulation scenario of multi-sen-sor multi-target tracking .     

Multi-sensor federated unscented Kalman filtering algorithm in intermittent observations

Aiming at the adverse effect caused by limited detecting probability of sensors on filtering precision of a nonlinear system state,a novel multi-sensor federated unscented Kalman filtering algorithm is proposed.Firstly,combined with the residual detection strategy,effective observations are correctly identified.Secondly,according to the missing characteristic of observations and the structural feature of unscented Kalman filter,the iterative process of the single-sensor unscented Kalman filter in intermittent observations is given.The key idea is that the state estimation and its error covariance matrix are replaced by the state one-step prediction and its error covariance matrix,when the phenomenon of observations missing occurs.Finally,based on the realization mechanism of federated filter,a new fusion framework of state estimation from each local node is designed.And the filtering precision of system state is improved further by the effective management of observations missing and the rational utilization of redundancy and complementary information among multi-sensor observations.The theory analysis and simulation results show the feasibility and effectiveness of the proposed algorithm.     

基于球面单径容积准则的多雷达分布式再入目标实时跟踪算法

针对多雷达对再入目标的实时跟踪问题,提出一种基于球面单径容积准则的分布式容积卡尔曼滤波算法。首先,利用球面单径容积准则近似计算非线性高斯权重积分,然后由统计线性误差传播方法等价表示滤波过程中的互协方差矩阵;最后通过一致性算法将单雷达容积卡尔曼滤波器所得时间更新结果与邻居雷达间进行信息交互与一致化处理,得到分布式球面单径容积卡尔曼滤波算法。该算法提高了再入弹道目标跟踪精度;无信息融合中心的通信拓扑结构降低了雷达间的通信量与计算量,提高了整个系统的生存能力。数值仿真结果验证了算法的有效性。     

Two-stage prediction and update particle filtering algorithm based on particle weight optimization in multi-sensor observation

The reasonable measuring of particle weight and effective sampling of particle state are consid- ered as two important aspects to obtain better estimation precision in particle filter. Aiming at the comprehensive treatment of above problems, a novel two-stage prediction and update particle filte- ring algorithm based on particle weight optimization in multi-sensor observation is proposed. Firstly, combined with the construction of muhi-senor observation likelihood function and the weight fusion principle, a new particle weight optimization strategy in multi-sensor observation is presented, and the reliability and stability of particle weight are improved by decreasing weight variance. In addi- tion, according to the prediction and update mechanism of particle filter and unscented Kalman fil- ter, a new realization of particle filter with two-stage prediction and update is given. The filter gain containing the latest observation information is used to directly optimize state estimation in the frame- work, which avoids a large calculation amount and the lack of universality in proposal distribution optimization way. The theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.     

Fuzzy Adaptive Strong Tracking Cubature Kalman Filter

To solve the problem that the choice of softening factor in conventional adaptive strong tracking filter( STF) greatly relies on the experience and computer simulation,a new concept of softening factor matrix is introduced and a fuzzy adaptive strong tracking cubature Kalman filter( FASTCKF) based on fuzzy logic controller is proposed. This method monitors residual absolute mean and standard deviation of each measurement component with fuzzy logic adaptive controller( FLAC),and adjusts the softening factor matrix dynamically by fuzzy rules,which is capable to modify suboptimal fading factor of STF adaptively and improve the filter's robust adaptive capacity. The simulation results show that the improved filtering performance is superior to the conventional square root cubature Kalman filter( SCKF) and the strong tracking square root cubature Kalman filter( STSCKF).     

未知噪声协方差的自适应容积卡尔曼滤波

针对噪声协方差不确定情况下容积卡尔曼滤波解决非线性目标跟踪中存在的问题,提出了一种优化的自适应容积卡尔曼滤波.首先根据新息序列和残差序列导出的线性矩阵方程得到噪声的协方差,基于新息序列与残差序列的相关性,推导出一种新的过程噪声协方差Q估计方法;然后采用残差序列对测量噪声协方差进行估计,利用加权因子将当前的噪声协方差矩阵与估计值组合成为新的测量噪声协方差阵R,有效避免了不准确状态估计的局限性.仿真结果表明:在时变噪声协方差的条件下,所提出的自适应容积卡尔曼算法的跟踪精度明显提高.     

闪烁噪声下目标跟踪的容积粒子滤波算法

针对闪烁噪声下非线性非高斯系统的目标跟踪问题,首先建立了闪烁噪声的数学模型;然后分析了传统粒子滤波算法的优劣点,在此基础上,引入容积卡尔曼滤波算法,重新设计粒子滤波的重要性密度函数,提出用容积粒子滤波算法来跟踪目标。最后进行了仿真分析与对比。仿真结果表明,闪烁噪声条件下,容积粒子滤波算法的跟踪误差分别是传统粒子滤波算法和无迹粒子滤波算法的1/5和1/2,有更高的跟踪精度;而运行时间仅是无迹粒子滤波算法的1/2,且跟踪稳定性更好。     

Federal extended Kalman filter based on reconstructed observation in incomplete observations

In the estimation and identification of nonlinear system state,aiming at the adverse effect of observation missing randomly caused by detection probability of used sensor which is less than 1,a novel federal extended Kalman filter( FEKF) based on reconstructed observation in incomplete observations( ROIO) is proposed in this paper. On the basis of multi-sensor observation sets,the observation is exchanged at different times to construct a new observation set. Based on each observation set,an extended Kalman filter algorithm is used to estimate the state of the target,and then the federal filtering algorithm is used to solve the state estimation based on the multi-sensor observation data. The effect of the sensor probing probability on the filtering result and the effect of the number of sensors on the filtering result are obtained by the simulation experiment,respectively. The simulation results demonstrate effectiveness of the proposed algorithm.     

粒子群优化平方根强跟踪CKF及应用

提出一种粒子群优化平方根强跟踪容积卡尔曼滤波算法,并将其用于水下应答器辅助航位推算组合导航系统. 以强跟踪滤波器为理论框架,结合容积卡尔曼滤波器,设计了平方根强跟踪容积卡尔曼滤波器. 提出一种改进的粒子群算法,将粒子两两为一对分成若干对,每进化一次后,比较两个粒子的代价函数值,代价函数值较优的粒子,搜索方向侧重于群体历史经验,代价函数较差的粒子,搜索方向侧重于自身历史经验. 将改进的粒子群算法用于求取强跟踪滤波器的渐消因子. 仿真结果表明在系统模型不准确的情况下所提算法依然能够有效跟踪状态变化,比传统的容积卡尔曼滤波器具有更高的滤波精度和稳定性.      

新型GPS动态定位自适应卡尔曼滤波方法

为了获取GPS卫星的信号及定位数据的真实值,减小信号传播中因各种因素混入的随机误差对定位精度的影响,通过应用运动载体"当前"统计模型,取速度和位置为观测量建立GPS动态定位模型,将观测量维数增大1倍,从而增加了系统的可观测性和定位测算精度.此外,针对传统标准卡尔曼滤波法在动态滤波方面的不足进行了分析,提出了改进型Sage自适应卡尔曼滤波法.该方法在递推和滤波过程中不断地修正模型参数,始终保持噪声模型接近于真实模型,从而避免了标准卡尔曼滤波法中因建模不准确可能导致的滤波发散等问题,较好地解决了GPS动态定位中状态变量维数与滤波快速性之间的矛盾,以及状态噪声和观测噪声建模不准确和时变的问题.     

基于GPS/INS组合导航的改进自适应渐消卡尔曼滤波算法

基于GPS/SINS组合导航系统的模型不准确或者量测噪声多变所产生的滤波发散问题,研究了自适应渐消卡尔曼滤波对于滤波发散的抑制作用,文章提出一种利用新息协方差估计值和量测值实时自适应计算渐消因子的方法,用它调节卡尔曼滤波方程中预测误差协方差阵和增益矩阵,调整历史新息和当前新息的权重达到抑制滤波发散的目的。该算法能有效减少严格收敛判据推导渐消因子的计算量和限制条件,有效利用了当前新息值。仿真验证表明,提出的算法能有效抑制滤波发散,并且比常规卡尔曼滤波效果更佳。     

基于改进扩展卡尔曼滤波的爬架运行安全姿态估计

为了实时准确的获取爬架运行时的姿态信息,提出了一种基于改进Sage-Husa扩展卡尔曼滤波算法（ISHEKF）的爬架姿态估计方法。首先建立了爬架姿态估计模型,然后在扩展卡尔曼滤波（EKF）对传感器进行融合滤波时,加入Sage-Husa扩展卡尔曼算法（SHEKF）对时变噪声进行调节,再以协方差匹配技术对Sage-Husa进行滤波发散判定,通过在调节因子中引入爬架实时运动速度,改进滤波发散判定依据,从而满足爬架运行时高精度的滤波要求。实验结果表明：静态实验中以横滚角为例,ISHEKF的最大误差较SHEKF减少了21.9%,较EKF最大误差减少了70.8%;动态实验中ISHEKF表现出更好的稳定性和滤波精度,能够准确反映爬架运行的状态变化。     

基于自适应滤波辅助残差χ2法的导航系统故障诊断方法

针对残差χ2法对软故障检测不敏感,且当量测噪声阵不准确时,也不能很好地检测突变故障的问题,提出了一种限制自适应滤波辅助残差法的诊断方法。基于贝叶斯理论推导出一种限制自适应滤波,利用其自适应调节性在线估计量测噪声阵,抑制滤波预报值跟踪软故障,以辅助残差χ2法进行故障检测。为了克服直接隔离故障法导致滤波精度降低,误检率增加的问题,提出了基于一步预测的故障处理和重构方法,即用前一时刻解算的结果预测当前时刻的导航信息。仿真结果表明,所提出的方法能实时有效地诊断出突变和软故障,保证系统在故障阶段的精度,并使系统在故障消失后及时恢复正常,提高了系统的可靠性。     

基于平方根UKF的水下纯方位目标跟踪

为了避免被动跟踪中非线性性带来的计算复杂化及跟踪精度的下降,该文将平方根无迹卡尔曼滤波(SR-UKF)算法应用到水下仅测角目标跟踪.利用协方差平方根代替协方差参加递推运算,解决了标准无迹卡尔曼滤波(UKF)算法中由于计算误差和噪声等因素有可能引起误差协方差矩阵负定而导致滤波结果发散的问题,保证了滤波算法的数值稳定性,提高了跟踪的精度和可靠性.仿真结果表明,SR-UKF非线性滤波算法应用于水下仅测角目标跟踪系统是有效的,而且滤波精度、稳定性和收敛时间明显优于扩展卡尔曼滤波(EKF)和标准UKF算法.