针对新息滤波交互式多模型(IF IMM)算法中切换过程模型概率滞后的问题,提出了模型概率转移矩阵马尔可夫参数自适应的新息滤波多模型算法(AM P-IF IMM),该方法采用后验信息修正不准确的先验信息,自适应的调整马尔可夫转移矩阵的参数.切换时刻较多地遗忘非匹配模型的信息,放大匹配模型的信息,在保证滤波精度的同时,大大提高了模型间切换速度.将该算法应用到CA,CV两模型组合导航系统取得了良好的效果. In order to solve the problem of probabilistic hysteresis in the switching process model of the IFIMM algorithm, a novel PMI algorithm, which is adaptive to the Markov parameters of model probability transfer matrix (AM P-IF IMM), is proposed The method adapts the a-posteriori information to correct the inaccurate prior information and adjusts the parameters of the Markov transition matrix adaptively. The information of the unmatched model is forgotten more at the moment of switching and the matching model information is amplified. At the same time, Which greatly improves the switching speed between models.This algorithm has been applied to CA and CV model combined navigation system and achieved good results.