针对被动传感器跟踪系统非线性较强问题,提出了一种基于改进高斯混合粒子滤波的被动传感器目标跟踪算法。 该算法基于Sigma点卡曼滤波和粒子滤波的特点,用有限的高斯混合模型来近似后验状态密度、系统噪声和观测噪声的 分布。然后结合遗传算法和EM算法来实现模型的降阶,克服了EM算法假定混合成分数为已知、迭代的结果需要依赖初 始值、可能收敛到局部最大点或可能收敛到参数空间的边界的缺点,从而改善粒子枯竭的问题。仿真实验结果表明在被 动传感器跟踪领域,与传统粒子滤波、基于EM的高斯混合粒子滤波和基于贪心EM的高斯混合粒子滤波相比,该算法在 保持高精度估计能力的同时,具有较强的鲁棒性,是解决非线性系统状态估计问题的一种有效方法。

An improved Gaussian mixture particle filter algorithm was proposed for the highly non-linear passive tracking system, the limited Gaussian mixture model was used to approximate the posterior density of states, system noise and measurement noise in the algorithm, which based on the characteristics of SPKF and particle filter. Then the genetic based EM algorithm was used to obtain the reduction of model order, which overcooked the disadvantage of the standard EM algorithm that assumed the number of the mixture components is a known priori, the performance of the overall parameter estimation process depends on the given good initial settings, and the estimated parameter can be resulted from some local optimum points. The effects caused by sampling depletion were lessened. Simulation results show that the algorithm outperforms the one based on PF, the one based on EM-GMPF and the one based on GEM-GMPF in tracking accuracy, and stability. Therefore it is more suitable to the nonlinear state estimation.