Multiple-Model Probability Hypothesis Density Filter for Tracking Maneuvering Targets

Tracking multiple targets with uncertain target dynamics is a difficult problem, especially with nonlinear state and/or measurement equations. With multiple targets, representing the full posterior distribution over target states is not practical. The problem becomes even more complicated when the number of targets varies, in which case the dimensionality of the state space itself becomes a discrete random variable. The probability hypothesis density (PHD) filter, which propagates only the first-order statistical moment (the PHD) of the full target posterior, has been shown to be a computationally efficient solution to multitarget tracking problems with a varying number of targets. The integral of PHD in any region of the state space gives the expected number of targets in that region. With maneuvering targets, detecting and tracking the changes in the target motion model also become important. The target dynamic model uncertainty can be resolved by assuming multiple models for possible motion modes and then combining the mode-dependent estimates in a manner similar to the one used in the interacting multiple model (IMM) estimator. This paper propose a multiple-model implementation of the PHD filter, which approximates the PHD by a set of weighted random samples propagated over time using sequential Monte Carlo (SMC) methods. The resulting filter can handle nonlinear, non-Gaussian dynamics with uncertain model parameters in multisensor-multitarget tracking scenarios. Simulation results are presented to show the effectiveness of the proposed filter over single-model PHD filters.     

MAP estimation of target manoeuvre sequence with theexpectation-maximization algorithm

Two algorithms are derived for the problem of tracking a manoeuvring target based on a sequence of noisy measurements of the state. Manoeuvres are modeled as unknown input (acceleration) terms entering linearly into the state equation and chosen from a discrete set. The expectation maximization (EM) algorithm is first applied, resulting in a multi-pass estimator of the MAP sequence of inputs. The expectation step for each pass involves computation of state estimates in a bank of Kalman smoothers tuned to the possible manoeuvre sequences. The maximization computation is efficiently implemented using the Viterbi algorithm. A second, recursive estimator is then derived using a modified EM-type cost function. To obtain a dynamic programming recursion, the target state is assumed to satisfy a Markov property with respect to the manoeuvre sequence. This results in a recursive but suboptimal estimator implementable on a Viterbi trellis. The transition costs of the latter algorithm, which depend on filtered estimates of the state, are compared with the costs arising in a Viterbi-based manoeuvre estimator due to Averbuch, et al. (1991). It is shown that the two criteria differ only in the weighting matrix of the quadratic part of the cost function. Simulations are provided to demonstrate the performance of both the batch and recursive estimators compared with Averbuch's method and the interacting multiple model filter     

Emitter localization using clustering-based bearing association

A closed-form emitter location estimator using time difference of arrival (TDOA) measurements is developed based on triangulation of hyperbolic asymptotes. The problem of associating the asymptotes with the emitter is solved by clustering the bearing angles of the linear asymptotes using a kernel density estimate. A closed-form estimate of the emitter location is obtained from triangulation of the clustered bearings using a weighted version of the pseudolinear estimator. By way of simulation examples, the proposed closed-form estimator is shown to outperform the computationally demanding and divergence-prone maximum likelihood (ML) estimator at moderate TDOA noise levels.     

Multi-EAP:Extended EAP for multi-estimate extraction for SMC-PHD filter

The ability to extract state-estimates for each target of a multi-target posterior, referred to as multi-estimate extraction (MEE), is an essential requirement for a multi-target filter, whose key performance assessments are based on accuracy, computational efficiency and reliability. The probability hypothesis density (PHD) filter, implemented by the sequential Monte Carlo approach, affords a computationally efficient solution to general multi-target filtering for a time-varying num-ber of targets, but leaves no clue for optimal MEE. In this paper, new data association techniques are proposed to distinguish real measurements of targets from clutter, as well as to associate par-ticles with measurements. The MEE problem is then formulated as a family of parallel single-estimate extraction problems, facilitating the use of the classic expected a posteriori (EAP) estima-tor, namely the multi-EAP (MEAP) estimator. The resulting MEAP estimator is free of iterative clustering computation, computes quickly and yields accurate and reliable estimates. Typical sim-ulation scenarios are employed to demonstrate the superiority of the MEAP estimator over existing methods in terms of faster processing speed and better estimation accuracy.     

论加速可靠性增长试验 (Ⅶ )步进应力试验方案的统计分析

对幂律-线性化模型步进应力加速可靠性增长试验方案，给出了模型参数，加速系数及MTBF的极大似然估计，极大后验估计，Bayes估计及其计算方法，对这些方案给出了Monte Carlo模拟方法，并用数值例说明了这些方法。     

Approximate Maximum Likelihood Algorithm for Moving Source Localization Using TDOA and FDOA Measurements

 A closed-form approximate maximum likelihood (AML) algorithm for estimating the position and velocity of a moving source is proposed by utilizing the time difference of arrival (TDOA) and frequency difference of arrival (FDOA) measurements of a signal received at a number of receivers. The maximum likelihood (ML) technique is a powerful tool to solve this problem. But a direct approach that uses the ML estimator to solve the localization problem is exhaustive search in the solution space, and it is very computationally expensive, and prohibits real-time processing. On the basis of ML function, a closed-form approximate solution to the ML equations can be obtained, which can allow real-time implementation as well as global convergence. Simulation results show that the proposed estimator achieves better performance than the two-step weighted least squares (WLS) approach, which makes it possible to attain the Cram閞-Rao lower bound (CRLB) at a sufficiently high noise level before the threshold effect occurs.     

具有丢包补偿的分布式一致性融合估计器

研究了具有随机丢包的网络化分布式一致性估计问题。丢包现象存在于各节点间局部状态估计值的传输过程中,引入一组服从Bernoulli分布的随机变量来描述。当发生丢包时,以融合节点前一时刻融合估计值的一步预测值进行补偿。建立了以估计器增益为决策变量,以所有传感器有限时域下状态融合估计误差和为代价函数的优化问题。在给定一致性权重下,通过最小化代价函数的上界得到了一组次优的估计器增益,并给出了融合估计器渐进稳定的充分条件。最后,通过算例仿真验证了算法的有效性。     

On state estimation for an orbiting single tether system

The effects of instrumentation accuracy and configuration on estimation error are studied for the small expandable-tether deployment system (SEDS) using a continuous-discrete extended Kalman filter (CDEKF) state estimator. A twelfth order model that incorporates the rigid body modes of the tether as well as the satellite attitude dynamics is developed. Simulation results using the model and the estimator indicate that the originally planned instrumentation package could not estimate the state vector adequately. Recommendations are made and results presented that reduce the estimation error by adding instruments and increasing selected measurement accuracies     

Orbital reference frame estimation with power spectral density constraints for drag-free satellites

The drag-free satellites are widely used in the field of fundamental science as they enable the high-precision measurement in pure gravity fields. This paper investigates the estimation of local orbital reference frame (LORF) for drag-free satellites. An approach, taking account of the combi-nation of the minimum estimation error and power spectral density (PSD) constraint in frequency domain, is proposed. Firstly, the relationship between eigenvalues of estimator and transfer func-tion is built to analyze the suppression and amplification effect on input signals and obtain the eigenvalue range. Secondly, an optimization model for state estimator design with minimum estima-tion error in time domain and PSD constraint in frequency domain is established. It is solved by the sequential quadratic programming (SQP) algorithm. Finally, the orbital reference frame estimation of low-earth-orbit satellite is taken as an example, and the estimator of minimum variance with PSD constraint is designed and analyzed using the method proposed in this paper.     

基于均值漂移无味粒子滤波的地面运动目标变结构交互多模型跟踪(英文)

为了跟踪地面运动目标,本文提出在变结构交互多模型基础上使用均值漂移无味粒子滤波的算法。模型滤波中,通过均值漂移将无味粒子滤波产生的采样粒子向目标状态最大后验密度估计方向移动。"停止"模型基础上,提出了"遮蔽"模型:出现地形遮蔽时,使用上一时刻的预测代替下一时刻的测量,且保持道路模型与遮蔽前一致。仿真实验采用地面运动目标指示雷达,考虑地面运动目标的三种常见场景:进入或离开道路、经过道路交叉点以及无测量值。使用了RMSE和ANEES两种评价指标,实验结果表明本文算法有效地提高了目标改变行驶道路时的跟踪精度;且目标速度过低或被遮蔽时,可以避免轨迹中断。     

基于混合系统粒子滤波和二元估计的故障预测算法

针对一类故障预测问题提出了一种基于粒子滤波的故障预测算法。在算法的状态估计阶段,采用混合系统粒子滤波和二元估计算法同时估计对象系统故障演化模型混合状态和未知参数的后验分布。在算法的状态预测阶段,在一定的假设条件的前提下,将混合模型连续状态变量的预测问题转化为一个基本状态空间模型的状态预测问题。通过对连续状态变量当前时刻的后验分布进行迭代采样从而获得其未来时刻的先验分布。在算法的决策阶段,在获取的故障演化模型连续状态变量分布基础上,结合一定的故障判据近似计算出对象系统剩余寿命分布。故障预测仿真实验结果证明了算法的有效性。     

Application of Road Information in Ground Moving Target Tracking

A new algorithm is developed to achieve accurate state estimation in ground moving target tracking by means of using road information. It is an adaptive variable structure interacting multiple model estimator with dynamic models modification （DMM VS-IMM for short）. Firstly, road information is employed to modify the target dynamic models used by filter, including modification of state transition matrix and process noise. Secondly, road information is applied to update the model set of a VS-IMM estimator. Predicted state estimation and road information are used to locate the target in the road network on which the model set is updated and finally IMM filtering is implemented. As compared with traditional methods, the accuracy of state estimation is improved for target moving not only on a single road, but also through an intersection. Monte Carlo simulation demonstrates the efficiency and robustness of the proposed algorithm with moderate computational loads.     

Recursive 3-D motion estimation from a monocular image sequence

Consideration is given to the design and application of a recursive algorithm to a sequence of images of a moving object to estimate both its structure and kinematics. The object is assumed to be rigid, and its motion is assumed to be smooth in the sense that it can be modeled by retaining an arbitrary number of terms in the appropriate Taylor series expansions. Translational motion involves a standard rectilinear model, while rotational motion is described with quaternions. Neglected terms of the Taylor series are modeled as process noise. A state-space model is constructed, incorporating both kinematic and structural states, and recursive techniques are used to estimate the state vector as a function of time. A set of object match points is assumed to be available. The problem is formulated as a parameter estimation and tracking problem which can use an arbitrarily large number of images in a sequence. The recursive estimation is done using an iterated extended Kalman filter (IEKF), initialized with the output of a batch algorithm run on the first few frames. Approximate Cramer-Rao lower bounds on the error covariance of the batch estimate are used as the initial state estimate error covariance of the IEKF. The performance of the recursive estimator is illustrated using both real and synthetic image sequences     

Adaptive detection threshold optimization for tracking in clutter

The adaptive optimization of detection thresholds for tracking in clutter is investigated for the probabilistic data association (PDA) filter. Earlier work on this problem by T.E. Fortmann et al. (1985) involved an approximate steady-state analysis of the state error covariance and is only suitable for time-invariant systems. Furthermore, the method requires numerous assumptions and approximations about the error covariance update equation, and uses a cumbersome graphical optimization algorithm. In this work we propose two adaptive schemes for threshold optimization, namely prior and posterior optimization algorithms which minimize the mean-square state estimation error over detection thresholds which depend on data up to the previous and current time-step, respectively. These algorithm are suitable for real-time implementation in time-varying systems. Some simulation results are presented     

Multistage integration model for human egomotion perception

Human computational vision models that attempt to account for the dynamic perception of egomotion and relative depth typically assume a common three-stage process: first, compute the optical flow field based on the dynamically changing image; second, estimate the egomotion states based on the flow; and third, estimate the relative depth/shape based on the egomotion states and possibly on a model of the viewed surface. We propose a model more in line with recent work in human vision, employing multistage integration. Here the dynamic image is first processed to generate spatial and temporal image gradients that drive a mutually interconnected state estimator and depth/shape estimator. The state estimator uses the image gradient information in combination with a depth/shape estimate of the viewed surface and an assumed model of the viewer's dynamics to generate current state estimates; in tandem, the depth/shape estimator uses the image gradient information in combination with the viewer's state estimate and assumed shape model to generate current depth/shape estimates. In this paper, we describe the model and compare model predictions with empirical data.     

电子整机加速贮存试验的Dirichlet分析方法

针对电子整机系统结构复杂,失效机理众多,无法利用传统的加速模型外推对其寿命和可靠性特征进行分析的问题,提出一种基于顺序Dirichlet分布的分析模型,利用多应力、多水平的环境应力,对每一阶段上的失效率建立指数分布模型。通过先验信息和基于反应论的修正加速模型,给出各应力水平上的失效率先验信息,利用多变量顺序Dirichlet分布描述先验失效率概率密度函数,并根据先验信息对Dirichlet分布参数进行辨识设计和对参数物理意义进行阐述。根据恒加定数试验特点,提出似然函数的解析步骤,利用Gibbs拒绝抽样方法对Dirichlet后验分布进行推断分析,得到后验信息。最后分析一个实例,给出抽样过程和几个分位点上的失效率估计值,并比较正常状态下先验和后验的可靠度变化趋势,验证算法具有一定的效率,为电子整机寿命预测与评估提供一种新方法。     

基于混合粒子滤波的激光惯导系统误差的补偿

针对激光陀螺捷联惯导系统在动态尤其是高动态环境下的姿态误差显著增大的问题,提出了一种基于改进高斯混合粒子滤波的纯方位跟踪算法。算法基于混合粒子的卡尔曼滤波和粒子滤波的特点,用有限的高斯模型来近似后验状态密度、系统噪声和观测噪声的分布通过EM的算法设计实现模型的降阶,一定程度上克服了EM算法迭代的结果需要依赖初始值、可能收敛到局部最大点或可能收敛到参数空间边界的缺点,从而改善了粒子携带信息的衰减问题。通过仿真与试验结合,在纯动态应用环境下的姿态与定位精度补偿效果,与传统Kalman滤波相比,算法在保持高精度估计能力的同时,具有较强的鲁棒性,是解决非线性系统状态估计问题的一种有效方法。     

State Estimation for Discrete Systems with Switching Parameters

The problem of state estimation for discrete systems with parameters which may be switching within a finite set of values is considered. In the general case it is shown that the optimal estimator requires a bank of elemental estimators with its number growing exponentially with time. For the Markov parameter case, it is found that the optimal estimator requires only N2 elemental estimators where N is the number of possible parameter values.