基于多速率交互式多模型的快速光电跟踪算法  

A Fast Optoelectronic Tracking Method Based on Multirate IMM Algorithm

作　　者：汪大宝[1] 刘上乾[1] 马彩文[2] 

机构地区：[1]西安电子科技大学技术物理学院,西安710071 [2]中国科学院西安光学精密机械研究所,西安710119

出　　处：《光子学报》2009年第12期3291-3294,共4页Acta Photonica Sinica

基　　金：国家自然科学基金(60377034)资助

摘　　要：针对交互式多模型粒子滤波算法运算量大的问题,提出了一种基于多速率跟踪思想的交互式多模型算法.该算法根据各模型假定的机动性,采用不同的数据更新速率,实现了模式空间和测量空间的混合滤波.同时,多模型综合选配了不同的滤波算法,其中,弱机动模型匹配卡尔曼滤波器,强机动模型匹配粒子滤波器.仿真结果表明,与传统的交互式多模型粒子滤波算法相比,本文算法在保证滤波精度的基础上,具有较低的计算复杂度,降低了约38.9%,能够有效地改善光电目标跟踪系统的可靠性和实时性.Aiming at disadvantages of heavy computational of interacting multiple model particle filter （IMMPF） algorithm,an interactive multiple model tracking algorithm based on muhirate tracking method （MRT） is proposed. The mixed filter between pattern space and measurement space is achieved,and the data is updated at a different rate according to the manoeuvrability of each model. And,each model marches different fiher： the poor maneuver model matches Kalman filter and the strong-maneuver model matches the particle filter. The simulation results show that the computational costs of this algorithm is reduced significantly by a factor of 38. 9% compared with that of traditional IMMPF algorithm,with comparable tracking performance to IMMPF. It can improve the real-time performance and reliability of optoelectronic target tracking system effectively.

关 键 词：机动目标跟踪 多速率跟踪 交互式多模型 粒子滤波 卡尔曼滤波 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]