基于IMM的UK-GMPHDF算法在多机动目标跟踪中的应用

为解决非线性系统滤波的非线性和多机动目标跟踪问题,提出了一种基于交互多模型(IMM)的无迹卡尔曼实现的高斯混合概率假设密度滤波(UK-GMPHDF)算法.该算法结合了IMM算法对不同目标机动模型的自适应能力和UK-GMPHD滤波精度高、计算量小的优点.此外,滤波器利用UK-GMPHD滤波,不仅避免了难以解决的数据关联问题,而且可以联合估计目标数和目标状态.在非线性系统和杂波环境下,通过对多机动目标跟踪的应用仿真,将该算法和基于单模型的UK-GMPHDF算法进行了比较,仿真结果表明了基于IMM的UK-GMPHDF算法具有较好的跟踪性能,大大提高了多机动目标跟踪精度,减少了跟踪的多目标误差.

Application of UK-GMPHDF algorithm based on IMM in multiple maneuvering targets tracking

For the purpose of solving the problem of nonlinear filtering and the multiple maneuvering targets tracking,a novel unscented Kalman implementations of the Gaussian mixture probability hypothesis density filter(UK-GMPHDF) based on interacting multiple model(IMM) is presented.The adaptive ability to various target maneuvering models is combined with the advantage of higher accuracy and lower computation load provided by UK-GMPHDF.Furthermore,the UK-GMPHDF avoids the data association problem and is able to jo...