高斯混合交互式多模型容积信息滤波算法

针对机动目标跟踪中由于目标机动使系统的非线性强度增大，导致系统的线性误差增大和跟踪精度明显下降、甚至发散的问题，提出了基于高斯混合的交互式多模型容积信息滤波（ GMIMM-CIF）算法，实现对机动目标的精确跟踪。新算法在每次输入交互之后，保留概率较大的几个假设，并利用一个高斯混合项替换最优多模型算法中剩余的假设，从而使算法中假设的数量保持恒定；用容积信息滤波器（ CIF）代替传统的非线性滤波器，通过估计信息状态向量和信息矩阵而不是估计状态向量和协方差，可以减小系统的非线性误差。通过仿真对比实验，验证了该算法可以提高机动目标的跟踪精度。

Gaussian mixture interacting multiple model algorithm based on cubature information filtering

Aiming at problem that increasing linear error and obvious decreasing in tracking precision and even diverging because of nonlinear strength increases when target maneuvers in the maneuvering target tracking system,a Gaussian mixture interacting multiple model algorithm based on cubature information filtering( GMIMM-CIF)is proposed to improve the tracking precision. The new algorithm utilizes a Gaussian mixture to replace the exponential number of hypotheses that arises in the optimal multiple model filtering at each step in each mode, thus maintaining a limited number of hypotheses;utilizes cubature information filtering to replace traditional nonlinear filtering,the cubature information filter( CIF)evaluates information state vector and information matrix rather than state vector and covariance,which can reduce the nonlinear error of system. Through simulation comparison experiment,verify that the proposed algorithm can improve tracking precision of maneuvering target.