强杂波环境下的LGM-PHDF算法  

LGM-PHD filter with strong clutter

作　　者：陈金广[1] 赵甜甜[1] 王明明[1] 马丽丽[1] CHEN Jin-guang;ZHAO Tian-tian;WANG Ming-ming;MA Li-li(School of Computer Science,Xi’an Polytechnic University,Xi’an 710048,China)

机构地区：[1]西安工程大学计算机科学学院,陕西西安710048

出　　处：《计算机工程与设计》2018年第2期532-536,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(61601358);陕西省自然科学基础研究计划基金项目(2016JM6030);中国纺织工业联合会科技指导性项目计划基金项目(2016074);陕西省教育厅科研计划基金项目(17JK0329);西安工程大学研究生创新基金项目(CX201631)

摘　　要：为解决LGM-PHDF算法在强杂波环境下,错误率和时间复杂度增加的问题,提出一种改进算法。在预测步骤结束后,使用量测信息和预测信息得到残差向量,通过椭球门限技术得到与目标真实状态接近的有效量测;在更新步骤中,只使用有效量测对高斯项进行更新,使用标签管理机制更新目标航迹。仿真结果表明,在强杂波环境中,改进算法降低了计算复杂度,具有更好的跟踪精度。To solve the problem that the complexity and the error rate of LGM-PHDF increase in strong clutter,an improved algorithm was proposed.The information at predicted step and measurements were used to calculate the residual vectors after prediction.Ellipsoid threshold technology was used to get the effective measurements which were near to the true target state.Only the effective measurements were used to update the Gaussian components.The mechanism of label management was used to update the target trajectories.Simulation results show that not only the complexity decreases,but also better accuracy is obtained using the improved algorithm.

关 键 词：强杂波 目标跟踪 带标签概率假设密度滤波 航迹关联 椭球门限 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]