摘要
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GMCBMeMBer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.
A multi-model cardinality balanced multi-target multi-Bernoulli （CBMeMBer） filter is proposed in this paper for tracking multiple maneuvering targets and forming the multi-target trajectories in clutter. Given the assumptions that the dynamic and observation models of the multi maneuvering targets are linear-Gaussian and by applying the Gaussian mixture （GM） technique, the analytic recursion for the proposed filter, namely the multi-model GM-CBMeMBer filter, is obtained. The extended Kalman （EK） filtering approximations for the nmlti-model GM-CBMeMBer filter to accommodate non-linear models are described briefly. Simulation results show that the proposed filter performs multiple maneuvering targets tracking well whereas the single-model GM-CBMeMBer filter obviously produces the missing and false trajectories. In addition, simulation results also show that for the scenarios of the relatively low signal-to-noise ratio （SNR）, the performance of the proposed filter is better than that of the multi-model GM probability hypothesis density （GM-PHD） filter, and is close to that of the multi-model GM cardinalized PHD （GM-CPHD） filter.
出处
《自动化学报》
EI
CSCD
北大核心
2014年第2期336-347,共12页
Acta Automatica Sinica
基金
国家重点基础研究发展计划(973计划)(2013CB329405)
国家自然科学基金创新研究群体(61221063)
中国博士后科学基金(20100481338)
中国博士后科学基金特别资助项目(2012T50746)
中央高校基本科研业务费专项资金资助~~
关键词
多机动目标跟踪
势平衡多目标多伯努利滤波器
交互式多模型算法
高斯混合实现
Multiple maneuvering targets tracking, cardinality balanced multi-target multi-Bernoulli （CBMeMBer） filter, interacting multiple models （IMM） algorithm, Gaussian mixture （GM） implementation
