基于ET-GM-PHD的机动多扩展目标跟踪算法

发布时间：2018-06-25 07:55

本文选题：多扩展目标 + 高斯混合概率假设密度　；参考：《计算机工程与应用》2017年17期

【摘要】：针对原始扩展目标高斯混合概率假设密度(Extended Target Gaussian Mixture Probability Hypothesis Density,ET-GM-PHD)滤波算法不能解决机动目标跟踪问题,在高斯混合概率假设密度(Gaussian Mixture Probability Hypothesis Density,GM-PHD)滤波框架下,引入修正的输入估计算法(Modified Input Estimation,MIE),可以有效地处理多扩展目标的机动问题。此外,提出的算法虽然可以实现对未知数目的多机动扩展目标进行跟踪,但无法获得各个目标的航迹。针对此问题,进一步引入高斯分量标记方法,有效地将多机动扩展目标的航迹进行准确关联,获取各个目标的航迹。实验结果表明,提出的算法在弱机动扩展目标跟踪中具有较好的跟踪性能,同时能够有效地估计多扩展目标的航迹。
[Abstract]:Aiming at the fact that the extended Target Gaussian mixture probability hypothesis (ET-GM-PHD) filtering algorithm can not solve the maneuvering target tracking problem, under the framework of Gao Si hybrid probability hypothesis density (GM-PHD) filtering, the proposed algorithm can not solve the maneuvering target tracking problem. The modified input estimation algorithm (mie) is introduced to deal with the maneuvering problem of multiple extended targets effectively. In addition, the proposed algorithm can track the unknown number of multi-maneuvering extended targets, but it can not obtain the track of each target. To solve this problem, the Gao Si component marking method is further introduced to effectively correlate the tracks of multiple maneuvering extended targets and obtain the tracks of each target. Experimental results show that the proposed algorithm has good tracking performance in weak maneuvering extended target tracking and can effectively estimate the track of multiple extended targets.
【作者单位】：江南大学物联网工程学院;江南大学轻工过程先进控制教育部重点实验室;
【基金】：国家自然科学基金(No.61305017) 江苏省自然科学基金(No.BK20130154) 江苏省普通高校研究生科研创新计划项目(No.KYLX15_1169)
【分类号】：TN713

【相似文献】