机动目标DOA跟踪粒子滤波算法
发布时间：2018-07-21 10:07
【摘要】：针对标准粒子滤波算法在机动目标波达方向(direction of arrival,DOA)随时间快速变化导致跟踪精度下降、实时性变差及多目标跟踪误差大等不足的问题,本文提出了一种改进粒子滤波(particle filter,PF)算法。该算法依据阵列信号处理模型和匀速(constant velocity,CV)模型,建立了机动目标跟踪的状态方程和观测方程作为状态空间模型,并在此基础上,借鉴多重信号分类(multiple signal classification,MUSIC)算法谱函数修改了粒子滤波的似然函数,实现了对目标方位的实时动态跟踪。仿真结果表明,与传统子空间类跟踪算法和标准粒子滤波算法相比,本文方法跟踪精度更高,收敛速度更快,抗噪能力及鲁棒性更强,对轨迹交叉的多目标跟踪性能也更优。
[Abstract]:In this paper, an improved particle filter (particle filter) algorithm is proposed to solve the problem that the fast change of the standard particle filter algorithm in the direction of arrival of maneuvering targets leads to the decrease of tracking accuracy, the real time variation and the large tracking error of multiple targets. In this paper, an improved particle filter (particle filter) algorithm is proposed. Based on array signal processing model and constant velocity CV model, the state equation and observation equation of maneuvering target tracking are established as state space model. The likelihood function of particle filter is modified by using the spectral function of (multiple signal classification music algorithm, and the real-time dynamic tracking of target azimuth is realized. The simulation results show that compared with the traditional subspace tracking algorithm and the standard particle filter algorithm, this method has higher tracking accuracy, faster convergence speed, stronger anti-noise ability and better robustness, and better performance for multi-target tracking with track crossing.
【作者单位】： 哈尔滨工程大学水声技术重点实验室;哈尔滨工程大学水声工程学院;解放军92985部队;
【基金】：国家自然科学基金项目(51279043);国家自然科学基金项目(61201411);国家自然科学基金项目(51209059) 国家“863”计划资助项目(2013AA09A503) 黑龙江省普通高校青年学术骨干支持计划(1253G019)
【分类号】：TN911.23
[Abstract]:In this paper, an improved particle filter (particle filter) algorithm is proposed to solve the problem that the fast change of the standard particle filter algorithm in the direction of arrival of maneuvering targets leads to the decrease of tracking accuracy, the real time variation and the large tracking error of multiple targets. In this paper, an improved particle filter (particle filter) algorithm is proposed. Based on array signal processing model and constant velocity CV model, the state equation and observation equation of maneuvering target tracking are established as state space model. The likelihood function of particle filter is modified by using the spectral function of (multiple signal classification music algorithm, and the real-time dynamic tracking of target azimuth is realized. The simulation results show that compared with the traditional subspace tracking algorithm and the standard particle filter algorithm, this method has higher tracking accuracy, faster convergence speed, stronger anti-noise ability and better robustness, and better performance for multi-target tracking with track crossing.
【作者单位】： 哈尔滨工程大学水声技术重点实验室;哈尔滨工程大学水声工程学院;解放军92985部队;
【基金】：国家自然科学基金项目(51279043);国家自然科学基金项目(61201411);国家自然科学基金项目(51209059) 国家“863”计划资助项目(2013AA09A503) 黑龙江省普通高校青年学术骨干支持计划(1253G019)
【分类号】：TN911.23
【相似文献】
相关期刊论文 前10条
1 夏佩伦;;潜艇对机动目标跟踪和攻击若干问题的探讨[J];火力与指挥控制;2010年02期
2 徐洪奎,王东进;基于卡尔曼滤波的组网雷达系统目标跟踪分析[J];系统工程与电子技术;2001年11期
3 陈玲,李少洪;基于无源时差定位系统的机动目标跟踪算法[J];系统工程与电子技术;2005年01期
4 夏艳林;梁雁冰;;电视与红外传感器对机动目标的分布式航迹融合[J];微计算机信息;2006年10期
5 周云锋;单甘霖;梅卫;;空中机动目标刚体运动建模方法[J];火力与指挥控制;2009年11期
6 张群,马长征,张守宏;机动目标三维成象中调频信号超分辨方法[J];应用科学学报;2000年02期
7 袁俊泉,皇甫堪;机动目标情况下FFT比相测距方法研究[J];现代雷达;2005年09期
8 王勇;姜义成;;两种线性时频分布在机动目标ISAR成像中的应用[J];哈尔滨工业大学学报;2006年05期
9 崔t熈，
本文编号：2135175
本文链接：https://www.wllwen.com/kejilunwen/wltx/2135175.html
