多目标无序量测OOSM-GMPHD滤波算法

摘要

For addressing multi-target tracking problem with out-of-sequence measurement (OOSM ),a filtering algorithm is proposed with multiple single-step-lag OOSM for linear system,which denotes as OOSM-Gaussian mixture probability hypothesis density (GMPHD ). Within the forward prediction framework,taking GMPHD as basis filtering algorithm,predicting and updating each Gaussian component,and then can obtained the target number and state estimation after pruning and merging. etc. Simulation results show that the algorithm can effectively filter out influence of OOSM and can accurately estimate the multi-target number and state.%针对无序量测条件下多目标跟踪问题,提出了一种适用于线性系统的单步滞后无序量测滤波算法(OOSM-GMPHD).在前向预测框架内,以高斯混合概率假设密度(GMPHD)滤波器为基础滤波算法,对每一高斯分量分别用延迟到达的量测与等价量测进行预测、更新,经剪枝与合并等步骤获得最终的目标数量与状态估计.仿真结果表明:算法可有效消除无序量测的影响,准确估计多目标数目和状态.