基于自适应马尔可夫参数交互多模型算法的弹道导弹跟踪研究

对弹道导弹主动段和自由段进行连续跟踪已成为弹道导弹防御亟待解决的问题。采用多模型交互算法，是解决该问题一条可行途径，但传统交互多模型马尔可夫概率转移矩阵参数固定，切换过程模型概率滞后。文中基于后验信息修正，提出了一种马尔可夫参数值自适应调整算法，根据不匹配模型误差压缩率的变化，自适应调整先验的马尔可夫概率转移矩阵参数，切换过程中较多地压缩不匹配模型的信息，放大匹配模型的信息，提高系统的收敛速度。仿真实验表明：马尔可夫参数自适应调整算法对弹道导弹主动段与自由段交替处的跟踪效果优于传统多模型交互算法。

Research on Ballistic Missile Tracking Based on Adaptive Markov Parameter IMM

How to consecutively track the boost phase and coast phase transitions without any prior information has become a pressing issue. Solving the problem with interacting multiple model is a feasible method. Since the Markov parameter of conventional interacting multiple model is constant, the model switching slows down. An adaptive Markov parameter IMM (AMP-IMM) algorithm is developed based on the modification of posterior information. The Markov transition probabilities can be modified adaptively in the process of filtering by omitting the information of non-matching model and magnifying the matching model information simultaneously during switching. The algorithm accelerates system convergence speed. Simulation results show that the proposed algorithm can consecutively track BM from the boost phase to the coast phase and is more effective than the conventional interacting multiple model.