基于改进的密度峰值聚类的扩展目标跟踪算法
Extended target tracking algorithm based on improved density peak clustering
-
摘要: 针对扩展目标高斯混合概率假设密度(extended target Gaussian mixture probability hypothesis density, ET-GM-PHD)滤波器中的量测集划分问题,提出了一种改进的密度峰值聚类(improved density peak clustering, IDPC)量测集划分算法。首先,使用IDPC算法去除局部密度较低的杂波量测,以获得最有可能的目标生成的量测集。其次,将剩余的量测集聚类以获得空间上紧密联系的聚类簇和簇的聚类中心。最后,根据预测的具有较高权重的高斯分量的均值在每个簇上的投影,获得准确的量测集划分。实验结果表明,与现有的量测集划分方法相比,该算法在保持跟踪精度的同时,可以大大减少计算时间。
Abstract: This paper proposes an improved density peak clustering (IDPC) measurement set partition algorithm to solve the problem of measurement set partition in the extended target Gaussian mixture probability hypothesis density (ET-GM-PHD) filter. Firstly, the IDPC algorithm removes the low local density clutter-generated measurements to obtain the most likely target-generated measurements. Secondly, cluster the remaining measurements is able to obtain spatially close clusters and cluster centers. Finally, according to the projection of the mean vectors of the predicted Gaussian component with higher weight on each cluster, an accurately measurement set partition is obtained. The simulation results show that, compared with the existing measurement set partition methods, this proposed algorithm can greatly reduce the calculation time while maintaining tracking accuracy.
下载: