模型组转换算法在工程应用中的改进研究

Improvement of Model Group Switching Algorithm in Engineering Application

在变结构多模估计当中,有一种算法称之为模型组转换算法,其可以用于跟踪运动规律较为复杂的目标。该文对此算法的模型组建立规则进行了改进,并将此算法与恒加速模型结合起来进行机动目标跟踪。仿真结果表明,改进的模型组建立规则有效的避免了滤波的发散,对其应用于目标跟踪起到了积极的推进作用。

A multiple-model （MM） estimator with a variable structure（VSMM） ,called model group switching（ MGS）algorithm, could track the targets with complicated movement. This paper brings forward an improved rule to build the model group, and the MGS algorithm is combined with CA model. Emulation indicates that the improved rule could avoid the infinite well, and it promotes the applying of the algorithm in the target tracking.