多目标跟踪的核粒子概率假设密度滤波算法

提出一种新的多目标跟踪算法：核粒子概率假设密度滤波算法(KP-PHDF)。算法的创新点在概率假设密度滤波算法(PHDF)的目标状态提取步骤,以粒子概率假设密度滤波算法为框架,并运用结合了mean-shift算法的核密度估计(KDE)理论进行概率假设密度(PHD)分布的二次估计、提取PHD峰值位置作为目标状态估计值。分析与多目标跟踪(MTT)仿真的结果表明,与现有序列蒙特卡罗概率假设密度滤波算法（SMC-PHDF）相比,在相同仿真条件下新算法的估计精度提高30.5%。     

混合线性/非线性模型的准高斯Rao-Blackwellized粒子滤波法

 针对混合线性/非线性模型,提出一种新的递推估计滤波算法,称为准高斯Rao-Blackwellized粒子滤波器（Q-GRBPF）。算法采用Rao-Blackwellized思想,将线性状态与非线性状态进行分离,对非线性状态运用准高斯粒子滤波（Q-GPF）算法进行估计,并将其后验分布近似为单个高斯分布,再利用非线性状态的估计值对线性状态进行卡尔曼滤波（KF）估计。将Q-GRBPF应用于目标跟踪的仿真结果表明,与Rao-Blackwellized粒子滤波器（RBPF）相比,Q-GRBPF在保证估计精度的前提下有效降低了计算复杂度,计算时间约为RBPF的58%;与Q-GPF相比,x坐标与y坐标的估计精度分别提升了45％和30％,而计算时间也节省了约30％。     

改进的MMPHD机动目标跟踪方法

基于序列蒙特卡罗方法的经典多模概率假设密度滤波方法及其各种衍生方法,在预测过程中依据多个并行的状态转移模型,通过将大量粒子散布到下一时刻目标所有可能出现的状态空间实现目标状态的捕获,造成计算量大、目标跟踪精度差。为此,提出一种改进的多模粒子概率假设密度机动目标跟踪方法。该方法利用最新量测信息估计目标运动模型概率及模型参数,并将估计得到的目标模型应用到粒子概率假设密度滤波方法的预测过程中生成预测粒子,从而将大部分粒子聚合在目标最可能出现的状态空间邻域中,实现粒子的有效利用。数值仿真表明,所提方法不仅显著地减少了目标丢失个数,而且提高了目标跟踪精度。     

基于CPHD的双层粒子滤波目标跟踪算法

常规基于势概率假设密度滤波(Cardinalized Probability Hypothesis Density,CPHD)的粒子滤波(Particle Fil? ter,PF)跟踪算法应用于多目标跟踪时,容易遇到因粒子数量增加而带来的运算效率下降、目标数目估计不准的问题。文章基于常规粒子滤波 CPHD跟踪算法,通过部署双层粒子,提出基于势概率假设密度滤波的双层粒子滤波 (Two-Layer Particle Filter-CPHD,TLPF-CPHD)算法,以便提高目标数目及状态估计精度。仿真实验结果证明,相比于常规 PF-CPHD算法,新算法具有更好的目标数目和状态估计准确性。     

交互多模型PHD滤波跟踪多目标方法

在机动多目标跟踪问题中,目标数未知或随时间而变化,概率假设密度（PHD）滤波可以在每一时间步估计多目标状态和目标数,但单模型方法不能给出精确的估计。提出了一种交互多模型PHD滤波方法,建立多模型描述多目标运动方式,利用PHD滤波结合多模型跟踪目标运动轨迹。同时,给出了多传感器交互多模型PHD滤波方法,以提高目标跟踪精度。     

区域杂波估计的多目标跟踪方法

高斯粒子概率假设密度（PHD）滤波往往假定杂波密度参数已知,这种做法对于实际应用是不现实的。此外,杂波的参数值通常依赖于环境条件,可能随时间发生变化。因此,多目标跟踪算法中需要实时准确估计杂波密度的参数。基于此,提出了一种多目标跟踪的区域杂波估计方法。首先根据量测信息在线估计出场景中的杂波数目,然后估计落入目标附近感兴趣区域的杂波数,并估计每个目标感兴趣区域杂波强度。仿真结果表明,在复杂场景下算法的跟踪性能明显优于未进行杂波估计的多目标跟踪算法,提高了跟踪的实时性和跟踪精度。     

未知测量噪声分布下的多目标跟踪算法

 粒子概率假设密度滤波(SMC-PHDF)在进行粒子更新时需要知道测量噪声的概率分布以计算似然函数,这使得SMC-PHDF依赖于测量噪声的概率模型。针对这一点不足,提出一种未知测量噪声分布下的多目标跟踪算法——基于风险评估的概率假设密度滤波(RE-PHDF)。该算法在SMC-PHDF进行概率假设密度(PHD)粒子更新时采用风险函数计算每个PHD粒子的风险值,并通过一个风险评估函数评估每个PHD粒子,然后用评估后的结果更新粒子的权值。由于粒子更新时避免了在多维测量空间中计算似然函数,算法不仅不依赖于测量噪声的概率分布,还可以节省大量计算时间。仿真结果表明：和SMC-PHDF相比,RE-PHDF在未知的复杂测量噪声环境下具有更高的鲁棒性和稳定性;同时,在两种算法跟踪精度接近的情况下,所提算法节省了50%的运行时间。     

幅度及多普勒信息辅助的多目标跟踪算法

针对现有的多目标跟踪（MTT）算法在强杂波环境下会出现严重的性能衰退,利用目标的幅度及多普勒特征,基于随机有限集（RFS）理论,提出了一种幅度及多普勒信息辅助的δ广义标签多伯努利（ADI-δ-GLMB）滤波器。首先,利用幅度及多普勒信息对目标状态进行扩展,在此基础上建立了新的量测似然函数。然后,基于δ-GLMB滤波器框架推导了新的更新方程。最后,进行仿真验证,结果表明：强杂波下,ADI-δ-GLMB滤波器跟踪性能明显优于δ-GLMB滤波器,估计精度及稳定性更高,且计算量更低;同信息辅助的概率假设密度（PHD）、多目标多伯努利（MeMBer）滤波器相比,ADI-δ-GLMB滤波器状态估计精度更高。     

高斯混合概率假设密度算法对多目标的跟踪研究

为了规避数据关联的困难,本文深入研究了适宜多目标跟踪工程应用,线性高斯多目标模型假设下的高斯混合概率假设密度算法（GM—PHD）,详细给出了后验PHD高斯元素的均值、方差和权值的解析递推式,使用了修剪和合并方法控制高斯元素数目的指数增长。最后。给出了一系列仿真实验,验证了在检测不确定和高杂波环境下,即使对目标数量未知和...     

混合线性/非线性状态空间模型的边缘Rao-Blackwellized粒子滤波法(英文)

本文提出了边缘 Rao-Blackwellized 粒子滤波器(marginal Rao-Blackwellized particle filter, MRBPF)算法,算法融合了 Rao-Blackwellized 粒子滤波器(Rao-Blackwellized particle filter , RBPF)算法和边缘粒子滤波器(marginal particle filter, MPF)算法。算法中状态被分为线形和非线性两部分,分别用 MPF 和卡尔曼滤波器(Kalman Filter)进行估计。地形辅助导航(terrain aided navigation, TAN)的仿真结果表明,与 RBPF 相比,提出算法的非线性状态估计的误差均方根(root mean square error, RMSE)和误差方差分别降低了约 29%和 96%,独立粒子数提高了约80%,获得了更好的收敛结果。分析表明,现有RBPF是MRBPF的一个特例。     

Track-before-detect for Infrared Maneuvering Dim Multi-target via MM-PHD

In this paper, we present a novel and efficient track-before-detect (TBD) algorithm based on multiple-model probability hypothesis density (MM-PHD) for tracking infrared maneuvering dim multi-target. Firstly, the standard sequential Monte Carlo probability hypothesis density (SMC-PHD) TBD-based algorithm is introduced and sequentially improved by the adaptive process noise and the importance re-sampling on particle likelihood, which result in the improvement in the algorithm robustness and convergence speed. Secondly, backward recursion of SMC-PHD is derived in order to ameliorate the tracking performance especially at the time of the multi-target arising. Finally, SMC-PHD is extended with multiple-model to track maneuvering dim multi-target. Extensive experiments have proved the efficiency of the presented algorithm in tracking infrared maneuvering dim multi-target, which produces better performance in track detection and tracking than other TBD-based algorithms including SMC-PHD, multiple-model particle filter (MM-PF), histogram probability multi-hypothesis tracking (H-PMHT) and Viterbi-like.     

The Marginal Rao-Blackwellized Particle Filter for Mixed Linear/Nonlinear State Space Models

In this paper, the marginal Rao-Blackwellized particle filter （MRBPF）, which fuses the Rao-Blackwellized particle filter （RBPF） algorithm and the marginal particle filter （MPF） algorithm, is presented. The state space is divided into linear and non-linear parts, which can be estimated separately by the MPF and the optional Kalman filter. Through simulation in the terrain aided navigation （TAN） domain, it is demonstrated that, compared with the RBPF, the root mean square errors （RMSE） and the error variance of the nonlinear state estimations by the proposed MRBPF are respectively reduced by 29% and 96%, while the unique particle count is increased by 80%. It is also found that the MRBPF has better convergence properties, and analysis has shown that the existing RBPF is nothing more than a special case of the MRBPF.     

非线性非高斯模型的高斯和PHD滤波算法(英文)

A new multi-target filtering algorithm, termed as the Gaussian sum probability hypothesis density （GSPHD） filter, is proposed for nonlinear non-Gaussian tracking models. Provided that the initial prior intensity of the states is Gaussian or can be identified as a Gaussian sum, the analytical results of the algorithm show that the posterior intensity at any subsequent time step remains a Gaussian sum under the assumption that the state noise, the measurement noise, target spawn intensity, new target birth intensity, target survival probability, and detection probability are all Gaussian sums. The analysis also shows that the existing Gaussian mixture probability hypothesis density （GMPHD） filter, which is unsuitable for handling the non-Gaussian noise cases, is no more than a special case of the proposed algorithm, which fills the shortage of incapability of treating non-Gaussian noise. The multi-target tracking simulation results verify the effectiveness of the proposed GSPHD.     

An improved multiple model GM-PHD filter for maneuvering target tracking

In this paper, an improved implementation of multiple model Gaussian mixture probability hypothesis density (MM-GM-PHD) filter is proposed. For maneuvering target tracking, based on joint distribution, the existing MM-GM-PHD filter is relatively complex. To simplify the filter, model conditioned distribution and model probability are used in the improved MM-GM-PHD filter. In the algorithm, every Gaussian components describing existing, birth and spawned targets are estimated by multiple model method. The final results of the Gaussian components are the fusion of multiple model estimations. The algorithm does not need to compute the joint PHD distribution and has a simpler computation procedure. Compared with single model GM-PHD, the algorithm gives more accurate estimation on the number and state of the targets. Compared with the existing MM-GM-PHD algorithm, it saves computation time by more than 30%. Moreover, it also outperforms the interacting multiple model joint probabilistic data association (IMMJPDA) filter in a relatively dense clutter environment.     

基于阵元激励幅度分档的赋形波束方向图综合

提出了一种新型的基于幅度分档的赋形波束方向图综合算法。该算法共分为3步。首先,使用传统方向图综合方法如交替投影得到波束的无幅值限制的阵元激励;然后,使用概率密度理论对得到的阵元激励幅度进行处理得到量化的阵元激励幅度,最后,通过量化阵元激励幅度,使用半正定松弛（SDR）方法得到阵元激励的相位分布。上述步骤中,如何使用概率密度理论得到量化的阵元激励幅度是3步中较为重要的一步。将阵元激励幅度用概率密度变量进行替代,通过事先设定的阵元激励幅度档位个数,以及每个阵元激励幅度落在相应档位时取值的概率,可以得到含有概率密度变量的综合方向图表达式。最小化含有概率密度变量的综合方向图与理想方向图的功率之差即可得到量化的阵元激励幅度。使用概率密度理论得到量化阵元激励幅度的优势在于,可以根据任意形状的阵面和阵元栅格排布来划分幅度的档位区间,从而有着更广泛的适用性。在例证部分,通过多组算例的仿真,以及与一些对算法性能的分析,所提算法验证了其在综合效果上的优越性。     

Multi-target state estimation and track continuity for the particle PHD filter

Particle filter approaches for approximating the first-order moment of a joint, or probability hypothesis density (PHD), have demonstrated a feasible suboptimal method for tracking a time-varying number of targets in real-time. We consider two techniques for estimating the target states at each iteration, namely k-means clustering and mixture modelling via the expectation-maximization (EM) algorithm. We present novel techniques for associating the targets between frames to enable track continuity.