数据采集与处理2011，Vol.26Issue(2)：181-187,7.
基于粒子滤波的交互式多模型多机动目标跟踪
Interacting Multiple Model Tracking Algorithm of Multiple Maneuvering Targets Based on Particle Filter
摘要
Abstract
The interacting multiple model joint probabilistic data association filtering (IMMJPDAF) algorithm has low tracking accuracy in nonlinar cases, and is inapplicabe for nonGaussion problem. Aiming at the disadvantages, an interacting multiple model target tracking algorithm of multiple maneuvering targets based on particle filter is proposed. The interacting multiple model joint probabilistic data association IMM-JPDA combined with particle filter (PF) is under the frame of IMM-JPDA. Every model uses PF to deal with the nonlinear and non-Gaussion problems, which avoid the Gausion assumption and the linearization error of nonlinear parts. Simulation results demonstrate that the tracking performance of IMM-JPDAPF algorithm is obviously superior to IMM-JPDAF algorithm. So the proposed algorithm can efficiently track multiple maneuvering targets in clutter circumstance.关键词
交互式多模型/联合概率数据关联/多目标跟踪/粒子滤波分类
信息技术与安全科学引用本文复制引用
章飞,周杏鹏,陈小惠..基于粒子滤波的交互式多模型多机动目标跟踪[J].数据采集与处理,2011,26(2):181-187,7.基金项目
海军装备预研基金资助项目. （）
