多模型粒子滤波跟踪算法研究

鉴福升; 徐跃民; 阴泽杰

doi:10.3724/SP.J.1146.2009.00853

多模型粒子滤波跟踪算法研究

doi: 10.3724/SP.J.1146.2009.00853

鉴福升^{①②③;徐跃民,},
徐跃民,
阴泽杰

计量
- 文章访问数: 3687
- HTML全文浏览量: 143
- PDF下载量: 1412
- 被引次数: 0
出版历程
- 收稿日期: 2009-06-05
- 修回日期: 2009-09-07
- 刊出日期: 2010-06-19

Research of Multiple Model Particle Filter Tracking Algorithms

Jian Fu-sheng^{①②③;徐跃民
,},
Xu Yue-min,
Yin Ze-jie

摘要

摘要: 针对机动目标跟踪问题，该文设计了一种改进的多模型粒子滤波(EMMPF)算法。与传统的多模型粒子滤波(MMPF)算法按照模型概率分配粒子数不同，该算法可根据用户定义的准则灵活控制各个模型的粒子数，且无需对模型间的粒子进行交互。模型估计和状态估计分开计算，并用模型似然函数更新模型后验概率。与MMPF进行的仿真比较表明，该算法能用较少的粒子数获得更好的滤波性能。
- 机动目标跟踪; 多模型; 粒子滤波; 粒子数
Abstract: An Enhanced Multiple Model Particle Filter(EMMPF) algorithm is presented for maneuvering target tracking. Rather than allocating the particles to the various modes according to mode probabilities as the MMPF, the new algorithm proposes an approach which enables the user to control the number of particles in a certain mode flexibly without interaction between particles of different mode. The estimations of mode and state are calculated respectively, and the posterior probability of each model is updated with the model likelihood function. It is demonstrated that the new algorithm can achieve better performance with less particles, compared with MMPF.

HTML全文

参考文献(1)

Blackman S S and Popoli R. Design and Analysis of Modern Tracking System [M]. Norwood MA: Artech House, 1999: 221-252.[2]Blom H A P and?Bloem E A. Exact Bayesian and particle filtering of stochastic hybrid systems[J].IEEE Transactions on Aerospace and Electronic Systems.2007, 43(1):55-70[3]Liang Yan, Wang Zeng-fu, and Cheng Yong-mei, et al.. Estimation of Markov jump systems with mode observation one-step lagged to state measurement [C]. The 10th International Conference on Information Fusion, Qubec City, Canada, 9-12 July 2007: 1-6.[4]Mcginnity S and Irwin G W. Multiple model bootstrap filter for maneuvering target tracking[J].IEEE Transactions on Aerospace and Electronic Systems.2000, 36(3):1006-1012[5]Driessen H and Boers Y. Efficient particle filter for jump Markov nonlinear systems[J].IEE Proceedings: Radar, Sonar and Navigation.2005, 152(5):323-326[6]Yacine M and Mohand S D. Genetic algorithm combined to IMM approach for tracking highly maneuvering targets[J]. IAENG International Journal of Computer Science, 2008, 35(1): 41-46.[7]刘贵喜, 高恩克, 范春宇. 改进的交互式多模型粒子滤波跟踪算法电子与信息学报[J]..2007, 29(12):2810-2813浏览Liu Gui-xi, Gao En-ke, and Fan Chun-yu. Tracking algorithms based on improved interacting multiple model particle filter[J].Journal of Electronics Information Technology.2007, 29(12):2810-2813[8]Doucet A, Gordon N, and Krishnamurthy V. Particle filters for state estimation of jump Markov linear systems[J].IEEE Transactions on Signal Processing.2001, 49(3):613-624[9]Caron F, Davy M, and Duflos E, et al.. Particle filtering for multisensor data fusion with switching observation models: Application to land vehicle positioning[J].IEEE Transactions on Signal Processing.2007, 55(6):2703-2719[10]Fredrik G, Niclas B, and Urban F, et al.. Particle filters for positioning, navigation and tracking[J].IEEE Transactions on Signal Processing.2002, 50(2):425-437

施引文献

资源附件(0)

访问统计