针对杂波、漏检等环境下的多机动目标估计问题,提出了两种非线性系统模型下多模型多伯努利(MM-MB)滤波器的实现方法.首先,基于容积卡尔曼滤波算法给出了MM-MB滤波器的容积卡尔曼高斯混合(CK-GM)实现过程;然后,为了增加CK-GM-MM-MB滤波算法的数值稳定性,给出了MM-MB滤波器的平方根容积卡尔曼高斯混合实现过程;最后,通过仿真实验验证了所提方法的有效性. To solve the problem of multi-maneuvering targets estimation under clutter and missed detection, two methods for the implementation of multi-model multi-Bernoulli (MM-MB) filters under nonlinear system model are proposed.Firstly, based on the volumetric Kalman filter The realization of volumetric Kalman-Gaussian mixture (CK-GM) of MM-MB filter is given. Then, in order to increase the numerical stability of CK-GM-MM-MB filtering algorithm, the square root of MM- Volume Kalman-Gaussian mixing process; Finally, the simulation results verify the effectiveness of the proposed method.