基于非对称交互多模型弹道导弹跟踪

弹道导弹飞行三个阶段的动力学模型异常复杂且非线性强，在未知导弹任何先验信息前提下如何连续地跟踪整个弹道已成为目前亟待解决的难题。针对该问题，提出了一种实时非对称交互多模型跟踪算法，它根据弹道不同阶段的受力情况建立相应的跟踪模型。该算法采用非对称的状态交互以及基于熵信息变化的模型概率更新方法对导弹进行实时状态估计，并将其中一个无迹滤波器的非线性状态估计用改进离散方法实现，提高了实时性。仿真实验表明，该跟踪算法能够较精确地从弹道主动段连续地跟踪到再入段，并且能较大幅度地减小模型切换之间的跟踪误差。

Ballistic missile tracking based on asymmetric interacting multiple model

The dynamical models of three phases of the ballistic missile (BM) trajectory are rather complicated and strongly nonlinear, how to consecutively track the whole BM trajectory without any prior information has become an urgent problem. To solve this problem, an asymmetric interacting multiple model (AIMM) filter is proposed, which establishes relevant tracking models based on different phases of the ballistic trajectory. The AIMM state interaction and a probability apdated method based on variations in entropy information interacting multiple model (IMM) are used to estimate the real time state of the BM. The algorithm also uses a discrete method to improve one of the unscented Kalman filters (UKF) nonlinear state estimation. Simulation results show that the proposed algorithm can consecutively track the BM from the boost phase till the reentry phase quite precisely; what is more, this algorithm can largely reduce the tracking error during the phase switching.