多模型GM-CBMeMBer滤波器及航迹形成

提出了一种可适用于杂波环境下对多个机动目标进行跟踪并能形成多目标航迹的多模型势平衡多目标多伯努利（Cardinality balanced multi-target multi-Bernoulli,CBMeMBer）滤波器. 随后,在多机动目标时间演化模型和观测模型均为线性高斯的假设条件下利用高斯混合（Gaussian mixture,GM）技术获得了该滤波器解析的递推形式——-多模型 GM-CBMeMBer 滤波器,并简要给出了它在非线性条件下的扩展卡尔曼（Extended Kalman,EK）滤波近似. 仿真实验结果表明所建议的多模型 GM-CBMeMBer 滤波器能有效地对多个机动目标进行跟踪而单模型 GM-CBMeMBer 滤波器则会产生明显的航迹丢失和虚假航迹,并且对于信噪比较低的仿真场景,它的性能优于多模型高斯混合概率假设密度（GM probability hypothesis density,GM-PHD）滤波器,接近于多模型高斯混合势概率假设密度（GM cardinalized PHD,GM-CPHD）滤波器.     

基于MMPHDF的多机动目标联合检测、跟踪与分类算法

针对场景中存在新目标出现、旧目标消失(即目标数目变化)和密集杂波的复杂情形,利用多模型概率假设密度滤波器(MMPHDF)在多机动目标联合检测与跟踪上的优势,加入类别辅助信息,提出了一种多机动目标联合检测、跟踪与分类算法.该算法的基本思想是在MMPHDF中用属性向量扩展单目标状态向量,用位置和属性的组合测量似然函数代替单目标位置及杂波位置测量似然函数,提高了不同类目标与杂波测量间的鉴别能力,从而改善了目标数目及状态的估计精度;在更新目标状态后,对目标属性信息进行更新,更为精确的目标数目及状态估计又保证了目标分类性能.本文给出了该算法的粒子实现方法.仿真结果验证了上述结论.     

On multiple-model extended target multi-Bernoulli filters

In this paper, we propose a multiple-model (MM) version of the extended target multi-Bernoulli (ET-MB) filter for estimating multiple maneuvering extended targets. A Gaussian mixture (GM) implementation of the MM-ET-MB filter for linear Gaussian models and a sequential Monte Carlo (SMC) implementation of the MM-ET-MB filter for nonlinear models are presented. Two numerical examples are provided to verify the effectiveness of the MM-ET-MB filter for estimating multiple maneuvering extended targets.     

机动多目标跟踪中的传感器控制策略的研究

针对机动多目标跟踪中的传感器控制问题, 本文提出一种基于信息论的多模型多伯努利滤波器的控制方案. 首先, 基于随机有限集(Random finite set, RFS)方法给出信息论下的传感器控制的一般方法; 其次, 本文给出多模型势均衡多目标多伯努利滤波器的序贯蒙特卡罗实现形式. 此外, 提出一种目标导向的多伯努利概率密度的粒子采样方法, 并借助该方法近似多目标概率密度, 继而利用Bhattacharyya 距离求解最终的控制方案. 典型机动多目标跟踪问题的仿真应用验证了本文传感器控制方法的有效性.     

基于GLMB滤波和Gibbs采样的多扩展目标有限混合建模与跟踪算法

本文针对杂波条件下多扩展目标的状态估计, 目标个数估计, 扩展目标形状估计问题, 提出了一种基于标签随机有限集(Labelled random finite sets, L-RFS)框架下多扩展目标跟踪学习算法, 该学习算法主要包括两方面:多扩展目标动态建模和多扩展目标的跟踪估计.首先, 结合广义标签多伯努利滤波器(Generalized labelled multi-Bernoulli, GLMB)建立了扩展目标的量测有限混合模型(Finite mixture models, FMM), 利用Gibbs采样和贝叶斯信息准则(Bayesian information criterion, BIC)准则推导出有限混合模型的参数来对多扩展目标形状进行学习, 然后采用等效量测方法来替代扩展目标产生的量测, 对扩展目标形状采用椭圆逼近建模, 实现扩展目标形状与状态的估计.仿真实验表明本文所给的方法能够有效跟踪多扩展目标, 并且在目标个数估计方面优于CBMeMBer算法.此外, 与标签多伯努利滤波(LMB)计算比较表明: GLMB和LMB算法滤波估计精度接近, 二者精度高于CBMeMBer算法.     

一种改进的广义标签多伯努利机动扩展目标跟踪算法

针对广义标签多伯努利滤波器(GLMB)预测步和更新步分别需要进行剪枝而导致计算量大、运行效率低且只考虑到单个运动模型的问题,提出一种多模型一步更新广义标签多伯努利机动扩展目标跟踪算法.首先通过公式推导将预测步与更新步合并,给出一种新的一步递归表达式;然后将多模型思想引入到一步递归表达式中,得到最终的多模型一步更新方程,同时基于吉布斯采样提出一种快速剪枝方法对其进行剪枝.由于改进后的滤波算法只涉及到一次剪枝且剪枝方法高效,算法的运行时间大大缩短;同时,由于采用了多模型思想,对机动目标的跟踪精度有了一定的提高.仿真实验表明,所提出的改进算法可以有效估计机动目标状态,且相比于多模型标签多伯努利滤波器(MMGLMB)计算效率明显提高.     

基于GGIW-PMB的衍生扩展目标跟踪

针对标准的扩展目标泊松多伯努利(Poisson multi-Bernoulli, PMB)滤波器难以有效跟踪衍生目标的问题,提出一种改进的PMB跟踪算法.算法采用随机矩阵法对扩展目标外形和尺寸建模,在滤波预测阶段利用多假设模型对衍生事件进行预测,得到多个伽玛高斯逆威沙特(gamma Gaussian inverse Wishart, GGIW)预测假设分量,最后在滤波更新阶段对预测分量更新得到扩展目标的运动状态和扩展形状估计.仿真结果表明,与标准的PMB滤波算法相比,所提算法有效改善衍生扩展目标的跟踪性能.     

基于ET-GM-PHD的机动多扩展目标跟踪算法

针对原始扩展目标高斯混合概率假设密度（Extended Target Gaussian Mixture Probability Hypothesis Density,ET-GM-PHD）滤波算法不能解决机动目标跟踪问题,在高斯混合概率假设密度（Gaussian Mixture Probability Hypothesis Density,GM-PHD）滤波框架下,引入修正的输入估计算法（Modified Input Estimation,MIE）,可以有效地处理多扩展目标的机动问题。此外,提出的算法虽然可以实现对未知数目的多机动扩展目标进行跟踪,但无法获得各个目标的航迹。针对此问题,进一步引入高斯分量标记方法,有效地将多机动扩展目标的航迹进行准确关联,获取各个目标的航迹。实验结果表明,提出的算法在弱机动扩展目标跟踪中具有较好的跟踪性能,同时能够有效地估计多扩展目标的航迹。     

Amultiplemaneuvering targets tracking algorithm based on a generalized pseudo-Bayesian estimator of first order

We describe the design of a multiple maneuvering targets tracking algorithm under the framework of Gaussian mixture probability hypothesis density (PHD) filter. First, a variation of the generalized pseudo-Bayesian estimator of first order (VGPB1) is designed to adapt to the Gaussian mixture PHD filter for jump Markov system models (JMS-PHD). The probability of each kinematic model, which is used in the JMS-PHD filter, is updated with VGPB1. The weighted sum of state, associated covariance, and weights for Gaussian components are then calculated. Pruning and merging techniques are also adopted in this algorithm to increase efficiency. Performance of the proposed algorithm is compared with that of the JMS-PHD filter. Monte-Carlo simulation results demonstrate that the optimal subpattern assignment (OSPA) distances of the proposed algorithm are lower than those of the JMS-PHD filter for maneuvering targets tracking.     

Probability hypothesis density filter based on strong tracking MIE for multiple maneuvering target tracking

Taking into account the difficulties of multiple maneuvering target tracking due to the unknown target number and the uncertain acceleration, a novel multiple maneuvering target tracking algorithm based on the Probability Hypothesis Density (PHD) filter and Modified Input Estimation (MIE) technique is proposed in this paper. First, the unknown acceleration vector is added to the target state to form a new augmented state vector. Then, strong tracking filter multiple fading factors are introduced to the MIE method which can adjust the prediction covariance and the corresponding filter gain at different rates in real time, so that the MIE method can adaptively track high maneuvering targets well. Finally, we combine this adaptive MIE method with the PHD filter, which can effectively track multiple maneuvering targets without much prior information. Simulation results show that the proposed algorithm has a higher tracking precision and a better real-time performance than the conventional maneuvering target tracking algorithms.     

基于星凸形随机超曲面模型多扩展目标多伯努利滤波器

针对复杂不确定性环境下具有不规则形状的多扩展目标跟踪问题, 提出了一种基于星凸形随机超曲面模型(Star-convex RHM)的多扩展目标多伯努利滤波算法.首先, 在有限集统计(Finite set statistics, FISST)理论框架下, 采用多伯努利随机有限集(MBer-RFS)和泊松RFS (Possion-RFS)分别描述多扩展目标的状态和观测, 并给出扩展目标势均衡多目标多伯努利(ET-CBMeMBer)滤波器.其次, 利用RHM去描述任意星凸形扩展目标的量测源分布, 提出了容积卡尔曼高斯混合星凸形多扩展目标多伯努利滤波器.此外, 本文给出了一种多扩展目标不规则形状估计性能的评价指标.最后, 通过多扩展目标和具有形状突变的多群目标的跟踪仿真实验验证了本文方法的有效性.     

基于分布式有限感知网络的多伯努利目标跟踪

针对感知范围受限的分布式传感网多目标跟踪问题, 在多伯努利滤波跟踪理论基础上提出分布式视场互补多伯努利关联算术平均融合跟踪方法. 首先, 通过视场互补扩大传感器感知范围, 其中, 局部公共区域只互补一次以降低计算成本. 其次, 每个传感器分别运行局部多伯努利滤波器, 并将滤波后验结果与相邻传感器进行泛洪通信使得每个传感器获取多个相邻传感器的后验信息. 随后, 通过距离划分进行多伯努利关联, 将对应于同一目标的伯努利分量关联到同一个子集中, 并对每个关联子集进行算术平均融合完成融合状态估计. 仿真实验表明, 所提方法在有限感知范围的分布式传感器网络中能有效地进行多目标跟踪.     

Fairness-based joint call admission control for heterogeneous wireless networks: an SMDP approach

In heterogeneous wireless networks, both terminal heterogeneity and network heterogeneity give rise to the fairness problem of resource allocation. Due to the capability of exploiting the resources of multiple networks, the behavior of multi-mode terminals will have a great effect on single-mode terminals, and this influence becomes more severe when considering the different demands of different traffic. In this article, we propose a novel joint call admission control （JCAC） scheme to address this problem. The JCAC problem is modeled as a semi-Markov decision process （SMDP） with the aim of maximizing the average network revenue under tile constraints of the fairness among different terminals and traffic classes. Based on the SMDP, we design an algorithm to achieve a good tradeoff between revenue and fairness by dynamically adjusting the threshold of fairness constraints imposed on heterogeneous terminals. Simulation results show that the proposed scheme can significantly improve the fairness among heterogeneous terminals and guarantee the priority and fairness among different traffic classes with little loss of network revenue compared with other schemes.     

基于箱粒子滤波的混合标签多伯努利跟踪算法

针对标签多伯努利滤波器在目标处于近邻或目标量测与轨迹关联模糊情况下,更新步中由于近似产生信息丢失,导致跟踪效果下降的问题,引入区间分析技术,结合标签多伯努利滤波器及广义标签多伯努利滤波器各自的优势,提出一种箱粒子滤波下的混合标签多伯努利跟踪算法.建立两种滤波器的参数模型,通过Kullback Leibler散度和熵两项评定标准在两种滤波器间进行切换,在特殊环境中使用广义标签多伯努利滤波器提高跟踪性能,在其他环境中使用标签多伯努利滤波器近似降低算法的复杂度,提高运算效率;同时基于箱粒子滤波实现混合标签多伯努利算法.仿真实验表明,在特定环境中,与原有滤波算法相比,所提出的改进算法在保证计算效率的同时,可提高跟踪的精确度及稳定性.     

强跟踪输入估计概率假设密度多机动目标跟踪算法

针对多机动目标跟踪中,目标数目未知及加速度不确定的问题,提出一种强跟踪输入估计（modifiedinputestimation,MIE）概率假设密度多机动目标跟踪算法．在详细分析算法的基础上,通过引入强跟踪多重渐消因子,以不同速率实时调节滤波器各个通道的预测协方差及相应的滤波器增益,从而实现MIE算法对加速度未知或发生人幅度突变的机动目标白适应跟踪能力;并将该算法与概率假设密度滤波算法有效结合,町以较好地跟踪未知数目的多机动目标．仿真结果表明,新算法比传统的多机动目标跟踪算法具有更岛的跟踪精度,且具有较好的实时性．     

Use of a kinematic constraint in tracking constant speed,maneuvering targets

The use of a kinematic constraint as a pseudomeasurement in the tracking of constant-speed, maneuvering targets is considered. The kinematic constraint provides additional information about the target motion that can be processed as a pseudomeasurement to improve tracking performances. A new formulation of the constraint equation is presented, and the rationale for the new formulation is discussed. The filter using the kinematic constraint as a pseudomeasurement is shown to be unbiased, and sufficient conditions for stochastic stability of the filter are given. Simulated tracking results are given to demonstrate the potential that the new formulation provides for improving tracking performance     

基于粒子滤波的交互式多模型多机动目标跟踪

针对交互式多模型联合概率数据关联滤波算法(IMM-JPDAF)在非线性情况下跟踪精度低,并不适用于非高斯问题的情况,提出了一种基于粒子滤波的交互式多模型多机动目标跟踪算法;将交互式多模型联合概率数据关联(IMM-JPDA)与粒子滤波相结合,在交互式多模型联合概率数据关联的框架下,各模型采用粒子滤波算法处理非线性非高斯问题,避免了噪声的高斯假设和非线性部分的线性化误差。仿真结果表明,IMM-JPDA-PF算法的跟踪性能明显优于IMM-JPDAF算法,能够对杂波环境中的多机动目标进行有效跟踪。     

Joint range ambiguity resolving and multiple maneuvering targets tracking in clutter via MMPHDF-DA

It is difficult to track multiple maneuvering targets of which the number is unknown and time- varying, especially when there is range ambiguity. The random finite sets （RFS） based probability hypothesis density filter （PHDF） is an effective solution to the problem of multiple targets tracking. However, when tracking multiple targets via the range ambiguous radar, the problem of range ambiguity has to be solved. In this paper, a multiple model PHDF and data association （MMPHDF-DA） based method is proposed to address multiple maneuvering targets tracking with range ambiguous radar in clutter. Firstly, by introducing the turn rate of target and the discrete pulse interval number （PIN） as components of target state vector, and modeling the incremental variable of the PIN as a three-state Markov chain, the problem of multiple maneuvering targets tracking with range ambiguity is converted into a hybrid state filtering problem. Then, by implementing a novel ＂track-estimate＂ oriented association with the filtering results of the hybrid filter, target tracks are provided at each time step. Simulation results demonstrate that the MMPHDF-DA can estimate target state as well as the PIN simultaneously, and succeeds in multiple maneuvering target tracking with range ambiguity in clutter. Simulation results also demonstrate that the MMPHDF-DA can overcome the limitation of the Chinese Remainder Theorem for range ambiguity resolving.     

Joint target tracking and classification with particle filtering and mixture Kalman filtering using kinematic radar information

This paper considers the problem of joint maneuvering target tracking and classification. Based on recently proposed Monte Carlo techniques, a multiple model (MM) particle filter and a mixture Kalman filter (MKF) are designed for two-class identification of air targets: commercial and military aircraft. The classification task is carried out by processing radar measurements only, no class (feature) measurements are used. A speed likelihood function for each class is defined using a prior information about speed constraints. Class-dependent speed likelihoods are calculated through the state estimates of each class-dependent tracker. They are combined with the kinematic measurement likelihoods in order to improve the classification process. The two designed estimators are compared and evaluated over rather complex target scenarios. The results demonstrate the usefulness of the proposed scheme for the incorporation of additional speed information. Both filters illustrate the opportunity of the particle filtering and mixture Kalman filtering to incorporate constraints in a natural way, providing reliable tracking and correct classification. Future observations contain valuable information about the current state of the dynamic systems. In the framework of the MKF, an algorithm for delayed estimation is designed for improving the current modal state estimate. It is used as an additional, more reliable information in resolving complicated classification situations.     

Distributed multirate interacting multiple model fusion (DMRIMMF) with application to out-of-sequence GMTI data

This note develops a distributed approach for fusing ground moving target indicator data with out-of-sequence (OOS) measurements. A multirate interacting multiple model (MRIMM) fusion algorithm is developed for effectively fusing multirate information. The multirate approach provides an excellent framework for efficient information retrodiction and forward update. A multirate interacting multiple model filter is employed locally to track a target with or without maneuvering behavior. The combination of global MRIMM fusion and local MRIMM tracking proves to be powerful for tracking and fusing maneuvering and nonmaneuvering targets in an environment of OOS measurement reporting.